CHAPTER 85: CHLAMYDIAL INFECTIONS

Chlamydiae are obligate intracellular bacteria that cause a wide variety of diseases in humans and animals.

The chlamydiae were originally classified as four species in the genus Chlamydia: C. trachomatis, C. pneumoniae, C. psittaci, and C. pecorum (the last species being found in ruminants). The C. psittaci group has been separated into three species: C. psittaci, C. felis, and C. abortus. The mouse pneumonitis strain (MoPn) is now classified as C. muridarum, and the guinea pig inclusion conjunctivitis strain (GPIC) is now designated C. caviae.

C. trachomatis is divided into two biovars: trachoma and LGV (lymphogranuloma venereum). The trachoma biovar causes two major types of disease in humans: ocular trachoma, the leading infectious cause of preventable blindness in the developing world; and urogenital infections, which are sexually or neonatally transmitted. The 18 serovars of C. trachomatis fall into three groups: the trachoma serovars A, B, Ba, and C; the oculogenital serovars D–K; and the LGV serovars L₁–L₃. Serovars can be distinguished by serologic typing with monoclonal antibodies or by molecular gene typing. However, serovar identification usually is not important clinically since the antibiotic susceptibility pattern is the same for all three groups. The one exception applies when LGV is suspected on clinical grounds; in this situation, serovar determination is important because a longer treatment duration is required for LGV strains.

BIOLOGY

During their intracellular growth, chlamydiae produce characteristic intracytoplasmic inclusions that can be visualized by direct fluorescent antibody (DFA) or Giemsa staining of infected clinical material, such as conjunctival scrapings or cervical or urethral epithelial cells. Chlamydiae are nonmotile, gram-negative, obligate intracellular bacteria that replicate within the cytoplasm of host cells, forming the characteristic membrane-bound inclusions that are the basis for some diagnostic tests. Originally considered to be large viruses, chlamydiae differ from viruses in possessing RNA and DNA as well as a cell wall that is quite similar in structure to the cell wall of typical gram-negative bacteria. However, chlamydiae lack peptidoglycan; their structural integrity depends on disulfide binding of outer-membrane proteins.

GROWTH CYCLE

Among the defining characteristics of chlamydiae is a unique growth cycle that involves alternation between two highly specialized morphologic forms (Figs. 85-1 and 85-2): the elementary body (EB), which is the infectious form and is specifically adapted for extracellular survival, and the metabolically active and replicating reticulate body (RB), which is not infectious, is adapted for an intracellular environment, and does not survive well outside the host cell. The biphasic growth cycle begins with attachment of the EB (diameter, 0.25–0.35 μm) at specific sites on the surface of the host cell. The EB enters the cell through a process similar to receptor-mediated endocytosis and resides in an inclusion, where the entire growth cycle is completed. The chlamydiae prevent phagosome-lysosome fusion. The inclusion membrane is modified by insertion of chlamydial antigens. Once the EB has entered the cell, it reorganizes into an RB, which is larger (0.5–1 μm) and contains more RNA. After ~8 h, the RB starts to divide by binary fission. The intracytoplasmic, membrane-bound inclusion body containing the RBs increases in size as the RBs multiply. Approximately 18–24 h after infection of the cell, these RBs begin to become EBs by a reorganization or condensation process that is poorly understood. After rupture of the inclusion body, the EBs are released to initiate another cycle of infection.

Chlamydiae are susceptible to many broad-spectrum antibiotics and possess a number of enzymes, but they have a very restricted metabolic capacity. None of these metabolic reactions results in the production of energy. Chlamydiae have thus been considered to be energy parasites that use the ATP produced by the host cell for their own metabolic functions. Many aspects of chlamydial molecular biology are not well understood, but the sequencing of several chlamydial genomes and new proteomics research have provided researchers with many relevant tools for elucidating the biology of the life cycle.

PATHOGENESIS

Genital infections are mostly caused by C. trachomatis serovars D–K, with serovars D, E, and F involved most often. Molecular typing of the major outer-membrane protein gene (omp1) from which serovar differences arise has been used to demonstrate that polymorphisms can occur in isolates from patients who are exposed frequently to multiple infections, while less variation is observed in isolates from less sexually active populations. Polymorphisms in the major outer-membrane protein may provide antigenic variation, and the different forms allow persistence in the community because immunity to one is not protective against the others.

The trachoma biovar is essentially a parasite of squamocolumnar epithelial cells; the LGV biovar is more invasive and involves lymphoid cells. As is typical of chlamydiae, C. trachomatis strains are capable of causing chronic, clinically inapparent, asymptomatic infections. Because the duration of the chlamydial growth cycle is ~48–72 h, the incubation period of sexually transmitted chlamydial infections is relatively long—generally 1–3 weeks. C. trachomatis causes cell death as a result of its replicative cycle and can induce cell damage whenever it persists. However, few toxic effects are demonstrated, and cell death because of chlamydial replication is not sufficient to account for disease manifestations, the majority of which are due to immunopathologic mechanisms or nonspecific host responses to the organism or its byproducts.

In recent years, the entire genomes of various chlamydial species have been sequenced, the field of proteomics has become established, host innate immunity has been more precisely delineated, and innovative host cell–chlamydial interaction studies have been conducted. As a result, many insights have been gained into how chlamydiae adapt and replicate in their intracellular environment and produce disease. These insights into pathogenesis include information on the regulation of gene expression, protein localization, the type III secretion system, the roles of CD4+ and CD8+ T lymphocytes in the host response, and T lymphocyte trafficking.

The chlamydial heat-shock protein, which shares antigenic epitopes with similar proteins of other bacteria and with human heat-shock protein, may sensitize the host, and repeated infections may cause host cell damage. Persistent or recurrent chlamydial infections are associated with fibrosis, scarring, and complications following simple epithelial infections. A common endpoint of these late consequences is scarring of mucous membranes. Genital complications can lead to pelvic inflammatory disease (PID) and its late consequences of infertility, ectopic pregnancy, and chronic pelvic pain, while ocular infections may lead to blinding trachoma. High levels of antibody to human heat-shock protein have been associated with tubal factor infertility and ectopic pregnancy. Without adequate therapy, chlamydial infections may persist for several years, although symptoms—if present—usually abate.

The pathogenic mechanisms of C. pneumoniae have yet to be completely elucidated. The same is true for C. psittaci, except that this agent infects cells very efficiently and causes disease that may reflect direct cytopathic effects.

GENITAL INFECTIONS

Spectrum

Although chlamydiae cause a number of human diseases, localized lower genital tract infections caused by C. trachomatis and the sequelae of such infections are the most important in terms of medical and economic impact. Oculogenital infections due to C. trachomatis serovars D–K are transmitted during sexual contact or from mother to baby during childbirth and are associated with many syndromes, including cervicitis, salpingitis, acute urethral syndrome, endometritis, ectopic pregnancy, infertility, and PID in female patients; urethritis, proctitis, and epididymitis in male patients; and conjunctivitis and pneumonia in infants. Women bear the greatest burden of morbidity because of the serious sequelae of these infections. Untreated infections lead to PID, and multiple episodes of PID can lead to tubal factor infertility and chronic pelvic pain. Studies estimate that up to 80–90% of women and >50% of men with C. trachomatis genital infections lack symptoms; other patients have very mild symptoms. Thus a large reservoir of infected persons continues to transmit infection to sexual partners.

As their designations reflect, the LGV serovars (L₁, L₂, and L₃) cause LGV, an invasive sexually transmitted disease (STD) characterized by acute lymphadenitis with bubo formation and/or acute hemorrhagic proctitis (see “Lymphogranuloma Venereum,” below).

Epidemiology

C. trachomatis genital infections are global in distribution. The World Health Organization (WHO) estimated in 2008 that >106.4 million cases occur annually worldwide. This figure makes chlamydial infection the most prevalent sexually transmitted bacterial infection in the world. The associated morbidity is substantial, and economic costs are high.

In the United States, chlamydial infections are the most commonly reported of all infectious diseases. In 2012, 1.3 million cases were reported to the U.S. Centers for Disease Control and Prevention (CDC); however, the CDC estimates that 2–3 million new cases occur per year, with substantial underreporting due to lack of screening in some populations. Rates of infection have increased every year; higher rates among women than among men reflect the focus on expansion of screening programs for women during the past 20 years, the use of increasingly sensitive diagnostic tests, an increased emphasis on case reporting, and improvements in the information systems used for reporting. The CDC and other professional organizations recommend annual screening of all sexually active women ≤25 years of age as well as rescreening of previously infected individuals at 3 months. Young women have the highest infection rates; in 2012, the figures were 3416.5 and 3722.5 cases per 100,000 population at 15–19 and 20–24 years of age, respectively. Age-specific rates among men, while much lower than those among women, were highest in the 20- to 24-year-old age group, at 1343.3 cases per 100,000. In 2012, rates increased for all racial and ethnic groups, with the highest rates among African Americans. For example, the rate of chlamydial infection among African-American girls 15–19 years of age was 7507.1 cases per 100,000—almost six times the rate among Caucasian girls in the same age group (1301.5/100,000). The rate among African-American women 20–24 years old was 4.8 times the rate among Caucasian women in the same age group. Similar racial disparities in reported rates of chlamydial infection exist among men. For boys 15–19 years of age, the rate among African Americans was 11.1 times the rate among Caucasians. The rate among Native Americans/Alaska Natives was more than four times the rate among Caucasians (648.3), and the rate among Latinos (383.6) was two times higher than that among Caucasians. These disparities are important reflections of health inequities in the United States.

The above statistics are based on case reporting. Studies based on screening surveys estimate that the U.S. prevalence of C. trachomatis cervical infection is 5% among asymptomatic female college students and prenatal patients, >10% among women seen in family planning clinics, and >20% among women seen in STD clinics. The prevalence of genital C. trachomatis infections varies substantially by geographic locale, with the highest rates in the southeastern United States. However, asymptomatic infections have been detected in >8–10% of young female military recruits from all parts of the country. The prevalence of C. trachomatis in the cervix of pregnant women is 5–10 times higher than that of Neisseria gonorrhoeae. The prevalence of genital infection with either agent is highest among women who are between the ages of 18 and 24, single, and non-Caucasian (e.g., African-American, Latina, Asian, Pacific Islander). Infections recur frequently in these same risk groups and are often acquired from untreated sexual partners. The use of oral contraception and the presence of cervical ectopy also confer an increased risk. The proportion of infections that are asymptomatic appears to be higher for C. trachomatis than for N. gonorrhoeae, and symptomatic C. trachomatis infections are clinically less severe. Mild or asymptomatic C. trachomatis infections of the fallopian tubes nonetheless cause ongoing tubal damage and infertility. The costs of C. trachomatis infections and their complications to the U.S. health care system have recently been estimated to exceed $516.7 million annually.

Clinical manifestations

Nongonococcal and postgonococcal urethritis

C. trachomatis is the most common cause of nongonococcal urethritis (NGU) and postgonococcal urethritis (PGU). The designation PGU refers to NGU developing in men 2–3 weeks after treatment of gonococcal urethritis with single doses of agents such as penicillin or cephalosporins, which lack antimicrobial activity against chlamydiae. Since current treatment regimens for gonorrhea have evolved and now include combination therapy with tetracycline, doxycycline, or azithromycin—all of which are effective against concomitant chlamydial infection—both the incidence of PGU and the causative role of C. trachomatis in this syndrome have declined.

In the United States, most of the estimated 2 million cases of acute urethritis are NGU, and C. trachomatis is implicated in 30–50% of these cases. The cause of most of the remaining cases of NGU is uncertain, but recent evidence suggests that Ureaplasma urealyticum, Mycoplasma genitalium, Trichomonas vaginalis, and herpes simplex virus (HSV) cause some cases. The rate of involvement of C. trachomatis in urethral infection ranges from 3–7% among asymptomatic men to 15–20% among symptomatic men attending STD clinics. A multisite study of men in Baltimore, Seattle, Denver, and San Francisco reported an overall chlamydial prevalence of 7% in urine samples assessed by nucleic acid amplification tests (NAATs). As in women, infection in men is age related, with young age as the greatest risk factor for chlamydial urethritis. The prevalence among men is highest at 20–24 years of age. In STD clinics, urethritis is usually less prevalent among men who have sex with men (MSM) than among heterosexual men and is almost always much more common among African-American men than among Caucasian men. One study reported prevalences of 19% and 9% among nonwhite and white heterosexual men, respectively.

NGU is diagnosed by documentation of a leukocytic urethral exudate and by exclusion of gonorrhea by Gram’s staining or culture. C. trachomatis urethritis is generally less severe than gonococcal urethritis, although in any individual patient these two forms of urethritis cannot reliably be differentiated solely on clinical grounds. Symptoms include urethral discharge (often whitish and mucoid rather than frankly purulent), dysuria, and urethral itching. Physical examination may reveal meatal erythema and tenderness as well as a urethral exudate that is often demonstrable only by stripping of the urethra.

At least one-third of male patients with C. trachomatis urethral infection have no evident signs or symptoms of urethritis. The availability of NAATs for first-void urine specimens has facilitated broader-based testing for asymptomatic infection in male patients. As a result, asymptomatic chlamydial urethritis has been demonstrated in 5–10% of sexually active male adolescents screened at school-based clinics or community centers. Such patients generally have pyuria (≥15 leukocytes per 400× microscopic field in the sediment of first-void urine), a positive leukocyte esterase test, or an increased number of leukocytes on a Gram-stained smear prepared from a urogenital swab inserted 1–2 cm into the anterior urethra. To differentiate between true urethritis and functional symptoms in symptomatic patients or to make a presumptive diagnosis of C. trachomatis infection in high-risk but asymptomatic men (e.g., male patients in STD clinics, sex partners of women with nongonococcal salpingitis or mucopurulent cervicitis, fathers of children with inclusion conjunctivitis), the examination of an endourethral specimen for increased leukocytes is useful if specific diagnostic tests for chlamydiae are not available. Alternatively, urethritis can be assayed noninvasively by examination of a first-void urine sample for pyuria, either by microscopy or by the leukocyte esterase test. Urine (or a urethral swab) can also be tested directly for chlamydiae by DNA amplification methods, as described below (see “Detection Methods”).

Epididymitis

Chlamydial urethritis may be followed by acute epididymitis, but this condition is rare, generally occurring in sexually active patients <35 years of age; in older men, epididymitis is usually associated with gram-negative bacterial infection and/or instrumentation procedures. It is estimated that 50–70% of cases of acute epididymitis are caused by C. trachomatis. The condition usually presents as unilateral scrotal pain with tenderness, swelling, and fever in a young man, often occurring in association with chlamydial urethritis. The illness may be mild enough to treat with oral antibiotics on an outpatient basis or severe enough to require hospitalization and parenteral therapy. Testicular torsion should be excluded promptly by radionuclide scan, Doppler flow study, or surgical exploration in a teenager or young adult who presents with acute unilateral testicular pain without urethritis. The possibility of testicular tumor or chronic infection (e.g., tuberculosis) should be excluded when a patient with unilateral intrascrotal pain and swelling does not respond to appropriate anti-microbial therapy.

Reactive arthritis

Reactive arthritis consists of conjunctivitis, urethritis (or, in female patients, cervicitis), arthritis, and characteristic mucocutaneous lesions. It may develop in 1–2% of cases of NGU and is thought to be the most common type of peripheral inflammatory arthritis in young men. C. trachomatis has been recovered from the urethra of 16–44% of patients with reactive arthritis and from 69% of men who have signs of urogenital inflammation at the time of examination. Antibodies to C. trachomatis have also been detected in 46–67% of patients with reactive arthritis, and Chlamydia-specific cell-mediated immunity has been documented in 72%. In addition, C. trachomatis has been isolated from synovial biopsy samples from 15 of 29 patients in a number of small series and from a smaller proportion of synovial fluid specimens. Chlamydial nucleic acids have been identified in synovial membranes and chlamydial EBs in joint fluid. The pathogenesis of reactive arthritis is unclear, but this condition probably represents an abnormal host response to a number of infectious agents, including those associated with bacterial gastroenteritis (e.g., Salmonella, Shigella, Yersinia, or Campylobacter), or to infection with C. trachomatis or N. gonorrhoeae. Since >80% of affected patients have the HLA-B27 phenotype and since other mucosal infections produce an identical syndrome, chlamydial infection is thought to initiate an aberrant hyperactive immune response that produces inflammation of the involved target organs in these genetically predisposed individuals. Evidence of exaggerated cell-mediated and humoral immune responses to chlamydial antigens in reactive arthritis supports this hypothesis. The finding of chlamydial EBs and DNA in joint fluid and synovial tissue from patients with reactive arthritis suggests that chlamydiae may actually spread from genital to joint tissues in these patients—perhaps in macrophages.

NGU is the initial manifestation of reactive arthritis in 80% of patients, typically occurring within 14 days after sexual exposure. The urethritis may be mild and may even go unnoticed by the patient. Similarly, gonococcal urethritis may precede reactive arthritis, but co-infection with an agent of NGU is difficult to rule out. The urethral discharge may be purulent or mucopurulent, and patients may or may not report dysuria. Accompanying prostatitis, usually asymptomatic, has been described. Arthritis usually begins ~4 weeks after the onset of urethritis but may develop sooner or, in a small percentage of cases, may actually precede urethritis. The knees are most frequently involved; next most commonly affected are the ankles and small joints of the feet. Sacroiliitis, either symmetrical or asymmetrical, is documented in two-thirds of patients. Mild bilateral conjunctivitis, iritis, keratitis, or uveitis is sometimes present but lasts for only a few days. Finally, dermatologic manifestations occur in up to 50% of patients. The initial lesions—usually papules with a central yellow spot—most often involve the soles and palms and, in ~25% of patients, eventually epithelialize and thicken to produce keratoderma blenorrhagicum. Circinate balanitis is usually painless and occurs in fewer than half of patients. The initial episode of reactive arthritis usually lasts 2–6 months.

Proctitis

Primary anal or rectal infections with C. trachomatis have been described in women and MSM who practice anal intercourse. In these infections, rectal involvement is initially characterized by severe anorectal pain, a bloody mucopurulent discharge, and tenesmus. Oculogenital serovars D–K and LGV serovars L₁, L₂, and L₃ have been found to cause proctitis. The LGV serovars are far more invasive and cause much more severely symptomatic disease, including severe ulcerative proctocolitis that can be clinically confused with HSV proctitis. Histologically, LGV proctitis may resemble Crohn’s disease in that giant cell formation and granulomas are detected. In the United States and Europe, cases of LGV proctitis occur almost exclusively in MSM, many of whom are positive for HIV infection.

The less invasive non-LGV serovars of C. trachomatis cause mild proctitis. Many infected individuals are asymptomatic, and in these cases infection is diagnosed only by routine culture or NAAT of rectal swabs. The number of fecal leukocytes is usually abnormal in both asymptomatic and symptomatic cases. Sigmoidoscopy may yield normal findings or may reveal mild inflammatory changes or small erosions or follicles in the lower 10 cm of the rectum. Histologic examination of rectal biopsies generally shows anal crypts and prominent follicles as well as neutrophilic infiltration of the lamina propria. Chlamydial proctitis is best diagnosed by isolation of C. trachomatis from the rectum and documentation of a response to appropriate therapy. NAATs are reportedly more sensitive than culture for diagnosis and are also specific.

Mucopurulent cervicitis

Although many women with chlamydial infections of the cervix have no symptoms, almost half generally have local signs of infection on examination. Cervicitis is usually characterized by the presence of a mucopurulent discharge, with >20 neutrophils per microscopic field visible in strands of cervical mucus in a thinly smeared, gram-stained preparation of endocervical exudate. Hypertrophic ectopy of the cervix may also be evident as an edematous area near the cervical os that is congested and bleeds easily on minor trauma (e.g., when a specimen is collected with a swab). A Papanicolaou smear shows increased numbers of neutrophils as well as a characteristic pattern of mononuclear inflammatory cells including plasma cells, transformed lymphocytes, and histiocytes. Cervical biopsy shows a predominantly mononuclear cell infiltrate of the subepithelial stroma. Clinical experience and collaborative studies indicate that a cutoff of >30 polymorphonuclear neutrophils (PMNs) per 1000× field in a gram-stained smear of cervical mucus correlates best with chlamydial or gonococcal cervicitis.

Clinical recognition of chlamydial cervicitis depends on a high index of suspicion and careful cervical examination. No genital symptoms are specifically correlated with chlamydial cervical infection. The differential diagnosis of a mucopurulent discharge from the endocervical canal in a young, sexually active woman includes gonococcal endocervicitis, salpingitis, endometritis, and intrauterine contraceptive device–induced inflammation. Diagnosis of cervicitis is based on the presence of PMNs on a cervical swab as noted above; the presence of chlamydiae is confirmed by either culture or NAAT.

Pelvic inflammatory disease

Inflammation of sections of the fallopian tube is often referred to as salpingitis or PID. The proportion of acute salpingitis cases caused by C. trachomatis varies geographically and with the population studied. It has been estimated that C. trachomatis causes up to 50% of PID cases in the United States. PID occurs via ascending intraluminal spread of C. trachomatis or N. gonorrhoeae from the lower genital tract. Mucopurulent cervicitis is often followed by endometritis, endosalpingitis, and finally pelvic peritonitis. Evidence of mucopurulent cervicitis is often found in women with laparoscopically verified salpingitis. Similarly, endometritis, demonstrated by an endometrial biopsy showing plasma cell infiltration of the endometrial epithelium, is documented in most women with laparoscopy-verified chlamydial (or gonococcal) salpingitis. Chlamydial endometritis can also occur in the absence of clinical evidence of salpingitis. Histologic evidence of endometritis has been correlated with a syndrome consisting of vaginal bleeding, lower abdominal pain, and uterine tenderness in the absence of adnexal tenderness. Chlamydial salpingitis produces milder symptoms than gonococcal salpingitis and may be associated with less marked adnexal tenderness. Thus, mild adnexal or uterine tenderness in a sexually active woman with cervicitis suggests chlamydial PID.

Chronic untreated endometrial and tubal inflammation can result in tubal scarring, impaired tubal function, tubal occlusion, and infertility, even among women who report no prior treatment for PID. C. trachomatis has been implicated particularly often in “subclinical” PID on the basis of (1) a lack of history of PID among Chlamydia-seropositive women with tubal damage or (2) detection of chlamydial DNA or antigen among asymptomatic women with tubal infertility. These data suggest that the best method to prevent PID and its sequelae is surveillance and control of lower genital tract infections along with diagnosis and treatment of sex partners and prevention of reinfections. Promotion of early symptom recognition and health care presentation may reduce the frequency and severity of sequelae of PID.

Perihepatitis

The Fitz-Hugh–Curtis syndrome was originally described as a complication of gonococcal PID. However, studies over the past several decades have suggested that chlamydial infection is more commonly associated with perihepatitis than is N. gonorrhoeae. Perihepatitis should be suspected in young, sexually active women who develop right-upper-quadrant pain, fever, or nausea. Evidence of salpingitis may or may not be found on examination. Frequently, perihepatitis is strongly associated with extensive tubal scarring, adhesions, and inflammation observed at laparoscopy, and high titers of antibody to the 57-kDa chlamydial heat-shock protein have been documented. Culture and/or serologic evidence of C. trachomatis is found in three-fourths of women with this syndrome.

Urethral syndrome in women

In the absence of infection with uropathogens such as coliforms or Staphylococcus saprophyticus, C. trachomatis is the pathogen most commonly isolated from college women with dysuria, frequency, and pyuria. Screening studies can recover C. trachomatis from both the cervix and the urethra; in up to 25% of infected women, the organism is isolated only from the urethra. The urethral syndrome in women consists of dysuria and frequency in conjunction with chlamydial urethritis, pyuria, and no bacteriuria or urinary pathogens. Although symptoms of the urethral syndrome may develop in some women with chlamydial infection, the majority of women attending STD clinics for urethral chlamydial infection do not have dysuria or frequency. Even in women with chlamydial urethritis causing the acute urethral syndrome, signs of urethritis such as urethral discharge, meatal redness, and swelling are uncommon. However, mucopurulent cervicitis in a woman presenting with dysuria and frequency strongly suggests C. trachomatis urethritis. Other correlates of chlamydial urethral syndrome include a duration of dysuria of >7–10 days, lack of hematuria, and lack of suprapubic tenderness. Abnormal urethral Gram’s stains showing >10 PMNs per 1000× field in women with dysuria but without coliform bacteriuria support the diagnosis of chlamydial urethritis. Other possible diagnoses include gonococcal or trichomonal infection of the urethra.

Infection in pregnancy and the neonatal period

Infections during pregnancy can be transmitted to infants during delivery. Approximately 20–30% of infants exposed to C. trachomatis in the birth canal develop conjunctivitis, and 10–15% subsequently develop pneumonia. Consequently, all newborn infants receive ocular prophylaxis at birth to prevent ophthalmia neonatorum. Without treatment, conjunctivitis usually develops at 5–19 days of life and often results in a profuse mucopurulent discharge. Roughly half of infected infants develop clinical evidence of inclusion conjunctivitis. However, it is impossible to differentiate chlamydial conjunctivitis from other forms of neonatal conjunctivitis (e.g., that due to N. gonorrhoeae, Haemophilus influenzae, Streptococcus pneumoniae, or HSV) on clinical grounds; thus laboratory diagnosis is required. Inclusions within epithelial cells are often detected in Giemsa-stained conjunctival smears, but these smears are considerably less sensitive than cultures or NAATs for chlamydiae. Gram-stained smears may show gonococci or occasional small gram-negative coccobacilli in Haemophilus conjunctivitis, but smears should be accompanied by cultures or NAATs for these agents.

C. trachomatis has also been isolated frequently and persistently from the nasopharynx, rectum, and vagina of infected infants—occasionally for >1 year in the absence of treatment. In some cases, otitis media results from perinatally acquired chlamydial infection. Pneumonia may develop in infants from 2 weeks to 4 months of age. C. trachomatis is estimated to cause 20–30% of pneumonia cases in infants <6 months of age. Epidemiologic studies have linked chlamydial pulmonary infection in infants with increased occurrence of subacute lung disease (bronchitis, asthma, wheezing) in later childhood.

Lymphogranuloma venereum

C. trachomatis serovars L₁, L₂, and L₃ cause LGV, an invasive systemic STD. The peak incidence of LGV corresponds with the age of greatest sexual activity: the second and third decades of life. The worldwide incidence of LGV is falling, but the disease is still endemic and a major cause of morbidity in parts of Asia, Africa, South America, and the Caribbean. LGV is rare in industrialized countries; for more than a decade, the reported incidence in the United States has been only 0.1 case per 100,000 population. In the Bahamas, an apparent outbreak of LGV was described in association with a concurrent increase in heterosexual infection with HIV. Reports of outbreaks with the newly identified variant L_2b in Europe, Australia, and the United States indicate that LGV is becoming more prevalent among MSM. These cases have usually presented as hemorrhagic proctocolitis in HIV-positive men. More widespread use of NAATs for identification of rectal infections may have enhanced case recognition.

LGV begins as a small painless papule that tends to ulcerate at the site of inoculation, often escaping attention. This primary lesion heals in a few days without scarring and is usually recognized as LGV only in retrospect. LGV strains of C. trachomatis have occasionally been recovered from genital ulcers and from the urethra of men and the endocervix of women who present with inguinal adenopathy; these areas may be the primary sites of infection in some cases. Proctitis is more common among people who practice receptive anal intercourse, and an elevated white blood cell count in anorectal smears may predict LGV in these patients. Ulcer formation may facilitate transmission of HIV infection and other sexually transmitted and blood-borne diseases.

As NAATs for C. trachomatis are being used more often, increasing numbers of cases of LGV proctitis are being recognized in MSM. Such patients present with anorectal pain and mucopurulent, bloody rectal discharge. Sigmoidoscopy reveals ulcerative proctitis or proctocolitis, with purulent exudate and mucosal bleeding. Histopathologic findings in the rectal mucosa include granulomas with giant cells, crypt abscesses, and extensive inflammation. These clinical, sigmoidoscopic, and histopathologic findings may closely resemble those of Crohn’s disease of the rectum.

The most common presenting picture in heterosexual men and women is the inguinal syndrome, which is characterized by painful inguinal lymphadenopathy beginning 2–6 weeks after presumed exposure; in rare instances, the onset comes after a few months. The inguinal adenopathy is unilateral in two-thirds of cases, and palpable enlargement of the iliac and femoral nodes is often evident on the same side as the enlarged inguinal nodes. The nodes are initially discrete, but progressive periadenitis results in a matted mass of nodes that becomes fluctuant and suppurative. The overlying skin becomes fixed, inflamed, and thin, and multiple draining fistulas finally develop. Extensive enlargement of chains of inguinal nodes above and below the inguinal ligament (“the sign of the groove”) is not specific and, although not uncommon, is documented in only a minority of cases. Spontaneous healing usually takes place after several months; inguinal scars or granulomatous masses of various sizes persist for life. Massive pelvic lymphadenopathy may lead to exploratory laparotomy.

Constitutional symptoms are common during the stage of regional lymphadenopathy and, in cases of proctitis, may include fever, chills, headache, meningismus, anorexia, myalgias, and arthralgias. Other systemic complications are infrequent but include arthritis with sterile effusion, aseptic meningitis, meningoencephalitis, conjunctivitis, hepatitis, and erythema nodosum (Fig. 14-40). Complications of untreated anorectal infection include perirectal abscess; anal fistulas; and rectovaginal, rectovesical, and ischiorectal fistulas. Secondary bacterial infection probably contributes to these complications. Rectal stricture is a late complication of anorectal infection and usually develops 2–6 cm from the anal orifice—i.e., at a site within reach on digital rectal examination. A small percentage of cases of LGV in men present as chronic progressive infiltrative, ulcerative, or fistular lesions of the penis, urethra, or scrotum. Associated lymphatic obstruction may produce elephantiasis. When urethral stricture occurs, it usually involves the posterior urethra and causes incontinence or difficulty with urination.

Diagnosis

Detection methods

Historically, chlamydiae were cultivated in the yolk sac of embryonated eggs. The organisms can be grown more easily in tissue culture, but cell culture—once considered the diagnostic gold standard—has been replaced by nonculture assays (Table 85-1). In general, culture for chlamydiae in clinical specimens is now performed only in specialized laboratories. The first nonculture assays, such as DFA staining of clinical material and enzyme immunoassay (EIA), have been replaced by NAATS, which are molecular tests that amplify the nucleic acids in clinical specimens. NAATS are currently recommended by the CDC as the diagnostic assays of choice; four or five NAAT assays approved by the U.S. Food and Drug Administration (FDA) are commercially available, some as high-throughput robotic platforms. Point-of-care diagnostic assays (including NAATs), by which patients can be treated before leaving the clinic, are of increasing interest and are becoming available.

Choice of specimen

Cervical and urethral swabs have traditionally been used for the diagnosis of STDs in female and male patients, respectively. However, given the greatly increased sensitivity and specificity of NAATs, less invasive samples (e.g., urine for both sexes and vaginal swabs for women) can be used. For screening of asymptomatic women, the CDC now recommends that self-collected or clinician-collected vaginal swabs, which are slightly more sensitive than urine, be used. Urine screening tests are often used in outreach screening programs, however. For symptomatic women undergoing a pelvic examination, cervical swab samples are desirable because they have slightly higher chlamydial counts. For male patients, a urine specimen is the sample of choice, but self-collected penile-meatal swabs have been explored.

Alternative specimen types

Ocular samples from babies and adults can be assessed by NAATs. However, since commercial NAATs for this purpose have not yet been approved by the FDA, laboratories must perform their own verification studies. Samples from rectal and pharyngeal sites have been used successfully to detect chlamydiae, but laboratories must verify test performance.

Other diagnostic issues

Because NAATs detect nucleic acids instead of live organisms, they should be used with caution as test-of-cure assays. Residual nucleic acid from cells rendered noninfective by antibiotics may continue to yield a positive result in NAATs until as long as 3 weeks after therapy, when viable organisms have actually been eradicated. Therefore, clinicians should not use NAATs for test of cure until after 3 weeks. The CDC currently does not recommend a test of cure after treatment for infection with C. trachomatis. However, because incidence studies have demonstrated that previous chlamydial infection increases the probability of becoming reinfected, the CDC does recommend that previously infected individuals be rescreened 3 months after treatment.

Serology

Serologic testing may be helpful in the diagnosis of LGV and neonatal pneumonia caused by C. trachomatis. The serologic test of choice is the microimmunofluorescence (MIF) test, in which high-titer purified EBs mixed with embryonated chicken yolk-sac material are affixed to a glass microscope slide to which dilutions of serum are applied. After incubation and washing, fluorescein-conjugated IgG or IgM antibody is applied. The test is read with an epifluorescence microscope, with the highest dilution of serum producing visible fluorescence designated as the titer. The MIF test is not widely available and is highly labor intensive. Although the complement fixation (CF) test also can be used, it employs only lipopolysaccharide (LPS) as the antigen and therefore identifies the pathogen only to the genus level. Single-point titers of >1:64 support a diagnosis of LGV, in which it is difficult to demonstrate rising antibody titers; i.e., paired serum samples are difficult to obtain since, by its very nature, the disease results in the patient’s being seen by the physician after the acute stage. Any antibody titer of >1:16 is considered significant evidence of exposure to chlamydiae. However, serologic testing is never recommended for diagnosis of uncomplicated genital infections of the cervix, urethra, and lower genital tract or for C. trachomatis screening of asymptomatic individuals.

TREATMENT

Prevention

Since many chlamydial infections are asymptomatic, effective control and prevention must involve periodic screening of individuals at risk. Selective cost-effective screening criteria have been developed. Among women, young age (generally <25 years) is a critical risk factor for chlamydial infections in nearly all studies. Other risk factors include mucopurulent cervicitis; multiple, new, or symptomatic male sex partners; and lack of barrier contraceptive use. In some settings, screening based on young age may be as sensitive as criteria that incorporate behavioral and clinical measures. Another strategy is universal testing of all patients in high-prevalence clinic populations (e.g., STD clinics, juvenile detention facilities, and family planning clinics).

The effectiveness of selective screening in reducing the prevalence of chlamydial infection among women has been demonstrated in several studies. In the Pacific Northwest, where extensive screening began in family planning clinics in 1998 and in STD clinics in 1993, the prevalence declined from 10% in the 1980s to <5% in 2000. Similar trends have occurred in association with screening programs elsewhere. In addition, screening can effect a reduction in upper genital tract disease. In Seattle, women at a large health maintenance organization who were screened for chlamydial infection on a routine basis had a lower incidence of symptomatic PID than did women who received standard care and underwent more selective screening.

In settings with low to moderate prevalence, the prevalence at which selective screening becomes more cost-effective than universal screening must be defined. Most studies have concluded that universal screening is preferable in settings with a chlamydial prevalence of >3–7%. Depending on the criteria used, selective screening is likely to be more cost-effective when prevalence falls below 3%. Nearly all regions of the United States have now initiated screening programs, particularly in family planning and STD clinics. Along with single-dose therapy, the availability of highly sensitive and specific diagnostic NAATs using urine specimens and self-obtained vaginal swabs makes it feasible to mount an effective nationwide Chlamydia control program, with screening of high-risk individuals in traditional health-care settings and in novel outreach and community-based settings. The U.S. Preventive Services Task Force has given C. trachomatis screening a Grade A recommendation, which means that private insurance and Medicare will cover its cost under the Affordable Care Act.

TRACHOMA

Epidemiology

Trachoma—a sequela of ocular disease in developing countries—continues to be a leading cause of preventable infectious blindness worldwide. The WHO estimates that ~6 million people have been blinded by trachoma and that ~1.3 million people in developing countries still suffer from preventable blindness due to trachoma; certainly hundreds of millions live in trachoma-endemic areas. Foci of trachoma persist in Australia, the South Pacific, and Latin America. Serovars A, B, Ba, and C are isolated from patients with clinical trachoma in areas of endemicity in developing countries in Africa, the Middle East, Asia, and South America.

The trachoma-hyperendemic areas of the world are in northern and sub-Saharan Africa, the Middle East, drier regions of the Indian subcontinent, and Southeast Asia. In hyperendemic areas, the prevalence of trachoma is essentially 100% by the second or third year of life. Active disease is most common among young children, who are the reservoir for trachoma. By adulthood, active infection is infrequent but sequelae result in blindness. In such areas, trachoma constitutes the major cause of blindness.

Trachoma is transmitted through contact with discharges from the eyes of infected patients. Transmission is most common under poor hygienic conditions and most often takes place between family members or between families with shared facilities. Flies can also transfer the mucopurulent ocular discharges, carrying the organisms on their legs from one person to another. The International Trachoma Initiative founded by the WHO in 1998 aims to eliminate blinding trachoma globally by 2020.

Clinical manifestations

Both endemic trachoma and adult inclusion conjunctivitis present initially as conjunctivitis characterized by small lymphoid follicles in the conjunctiva. In regions with hyperendemic classic blinding trachoma, the disease usually starts insidiously before the age of 2 years. Reinfection is common and probably contributes to the pathogenesis of trachoma. Studies using polymerase chain reaction (PCR) or other NAATs indicate that chlamydial DNA is often present in the ocular secretions of patients with trachoma, even in the absence of positive cultures. Thus persistent infection may be more common than was previously thought.

The cornea becomes involved, with inflammatory leukocytic infiltrations and superficial vascularization (pannus formation). As the inflammation continues, conjunctival scarring eventually distorts the eyelids, causing them to turn inward so that the lashes constantly abrade the eyeball (trichiasis and entropion); eventually the corneal epithelium is abraded and may ulcerate, with subsequent corneal scarring and blindness. Destruction of the conjunctival goblet cells, lacrimal ducts, and lacrimal gland may produce a “dry-eye” syndrome, with resultant corneal opacity due to drying (xerosis) or secondary bacterial corneal ulcers.

Communities with blinding trachoma often experience seasonal epidemics of conjunctivitis due to H. influenzae that contribute to the intensity of the inflammatory process. In such areas, the active infectious process usually resolves spontaneously in affected persons at 10–15 years of age, but conjunctival scars continue to shrink, producing trichiasis and entropion with subsequent corneal scarring in adults. In areas with milder and less prevalent disease, the process may be much slower, with active disease continuing into adulthood; blindness is rare in these cases.

Eye infection with oculogenital C. trachomatis strains in sexually active young adults presents as an acute onset of unilateral follicular conjunctivitis and preauricular lymphadenopathy similar to that seen in acute conjunctivitis caused by adenovirus or HSV. If untreated, the disease may persist for 6 weeks to 2 years. It is frequently associated with corneal inflammation in the form of discrete opacities (“infiltrates”), punctate epithelial erosions, and minor degrees of superficial corneal vascularization. Very rarely, conjunctival scarring and eyelid distortion occur, particularly in patients treated for many months with topical glucocorticoids. Recurrent eye infections develop most often in patients whose sexual partners are not treated with antimicrobial agents.

Diagnosis

The clinical diagnosis of classic trachoma can be made if two of the following signs are present: (1) lymphoid follicles on the upper tarsal conjunctiva; (2) typical conjunctival scarring; (3) vascular pannus; or (4) limbal follicles or their sequelae, Herbert’s pits. The clinical diagnosis of endemic trachoma should be confirmed by laboratory tests in children with relatively marked degrees of inflammation. Intracytoplasmic chlamydial inclusions are found in 10–60% of Giemsa-stained conjunctival smears in such populations, but chlamydial NAATs are more sensitive and are often positive when smears or cultures are negative. Follicular conjunctivitis in European or American adults living in trachomatous regions is rarely due to trachoma.

TREATMENT

Psittacine birds and many other avian species act as natural reservoirs for C. psittaci–type organisms, which are common pathogens in domestic mammals and birds. The species C. psittaci, which now includes only avian strains, affects humans only as a zoonosis. (The other strains previously included in this species have been placed into different species that generally reflect the animals they infect: C. abortus, C. muridarum, C. suis, C. felis, and C. caviae.) Although all birds are susceptible, pet birds (parrots, parakeets, macaws, and cockatiels) and poultry (turkeys and ducks) are most frequently involved in transmission of C. psittaci to humans. Exposure is greatest among poultry-processing workers and owners of pet birds. Infectious forms of the organisms are shed from both symptomatic and apparently healthy birds and may remain viable for several months. C. psittaci can be transmitted to humans by direct contact with infected birds or by inhalation of aerosols from avian nasal discharges and from infectious avian fecal or feather dust. Transmission from person to person has never been demonstrated.

The diagnosis is usually established serologically. Psittacosis in humans may present as acute primary atypical pneumonia (which can be fatal in up to 10% of untreated cases); as severe chronic pneumonia; or as a mild illness or asymptomatic infection in persons exposed to infected birds.

EPIDEMIOLOGY

Fewer than 50 confirmed cases of psittacosis are reported in the United States each year, although many more cases probably occur than are reported. Control of psittacosis depends on control of avian sources of infection. A pandemic of psittacosis was once stopped by banning shipment or importation of psittacine birds. Birds can receive prophylaxis in the form of a tetracycline-containing feed. Imported birds are currently quarantined for 30 days of treatment.

CLINICAL MANIFESTATIONS

Typical symptoms include fever, chills, muscular aches and pains, severe headache, hepato- and/or splenomegaly, and gastrointestinal symptoms. Cardiac complications may involve endocarditis and myocarditis. Fatal cases were common in the preantibiotic era. As a result of quarantine of imported birds and improved veterinary-hygienic measures, outbreaks and sporadic cases of psittacosis are now rare. Severe pneumonia requiring management in an intensive care unit may develop. Endocarditis, hepatitis, and neurologic complications may occur, and fatal cases have been reported. The incubation period is usually 5–19 days but can last as long as 28 days.

DIAGNOSIS

Previously, the most widely used serologic test for diagnosing chlamydial infections was the genus-specific CF test, in which assay of paired serum specimens often shows fourfold or greater increases in antibody titer. The CF test remains useful, but the gold standard of serologic tests is now the MIF test, which is not widely available (see section on diagnosis of C. trachomatis genital infection, above). Any antibody titer above 1:16 is considered significant evidence of exposure to chlamydiae, and a fourfold titer rise in paired sera in combination with a clinically compatible syndrome can be used to diagnose psittacosis. Some commercially available serologic tests based on measurement of antibodies to LPS can be useful when the clinical diagnosis is consistent with bird exposure; however, since these tests are reactive for all chlamydiae (i.e., all chlamydiae contain LPS), caution must be used in their interpretation.

TREATMENT

C. pneumoniae is a common cause of human respiratory diseases, such as pneumonia and bronchitis. This organism has been reported to account for as many as 10% of cases of community-acquired pneumonia, most of which are diagnosed by serology. Serologic studies have linked C. pneumoniae to atherosclerosis; isolation and PCR detection in cardiovascular tissues have also been reported. These findings suggest an expanded range of diseases and syndromes for C. pneumoniae. The role of C. pneumoniae in the etiology of atherosclerosis has been discussed since 1988, when Finnish researchers presented serologic evidence of an association of this organism with coronary heart disease and acute myocardial infarction. Subsequently, the organism was identified in atherosclerotic lesions by culture, PCR, immunohistochemistry, and transmission electron microscopy; however, discrepant study results (including those of animal studies) and failure of large-scale treatment studies have raised doubts as to the etiologic role of C. pneumoniae in atherosclerosis. Large-scale case-cohort studies have demonstrated some association of C. pneumoniae with lung cancer, as evaluated by serology.

EPIDEMIOLOGY

Primary infection occurs mainly in school-aged children and reinfection in adults. Seroprevalence rates of 40–70% show that C. pneumoniae is widespread in both industrialized and developing countries. Seropositivity usually is first detected at school age, and rates generally increase by ~10% per decade. About 50% of individuals have detectable antibody at 30 years of age, and most have detectable antibody by the eighth decade of life. Although serologic evidence suggests that C. pneumoniae may be associated with up to 10% of cases of community-acquired pneumonia, most of this evidence is based not on paired serum samples but rather on a single high IgG titer. Some doubt exists about the true prevalence and etiologic role of C. pneumoniae in atypical pneumonia, especially since reports of cross-reactivity have raised questions about the specificity of serology when only a single serum sample is used for diagnosis.

PATHOGENESIS

Little is known about the pathogenesis of C. pneumoniae infection. It begins in the upper respiratory tract and, in many persons, persists as a prolonged asymptomatic condition of the upper respiratory mucosal surfaces. However, evidence of replication within vascular endothelium and synovial membranes of joints shows that, in at least some individuals, the organism is transported to distant sites, perhaps within macrophages. A C. pneumoniae outer-membrane protein may induce host immune responses whose cross-reactivity with human proteins results in an autoimmune reaction.

As mentioned above, epidemiologic studies have demonstrated an association between serologic evidence of C. pneumoniae infection and atherosclerotic disease of the coronary and other arteries. In addition, C. pneumoniae has been identified in atherosclerotic plaques by electron microscopy, DNA hybridization, and immunocytochemistry. The organism has been recovered in culture from atheromatous plaque—a result indicating the presence of viable replicating bacteria in vessels. Evidence from animal models supports the hypothesis that C. pneumoniae infection of the upper respiratory tract is followed by recovery of the organism from atheromatous lesions in the aorta and that the infection accelerates the process of atherosclerosis, especially in hypercholesterolemic animals. Antimicrobial treatment of the infected animals reverses the increased risk of atherosclerosis. In humans, two small trials in patients with unstable angina or recent myocardial infarction suggested that antibiotics reduce the likelihood of subsequent untoward cardiac events. However, larger-scale trials have not documented an effect of various antichlamydial regimens on the risk of these events.

CLINICAL MANIFESTATIONS

C. pneumoniae was first reported as the etiologic agent of mild atypical pneumonia in military recruits and college students. The clinical spectrum of C. pneumoniae infection includes acute pharyngitis, sinusitis, bronchitis, and pneumonitis, primarily in young adults. The clinical manifestations of primary infection appear to be more severe and prolonged than those of reinfection. The pneumonitis of C. pneumoniae pneumonia resembles that of Mycoplasma pneumonia in that leukocytosis is frequently lacking and patients often have prominent antecedent upper respiratory tract symptoms, fever, nonproductive cough, mild to moderate illness, minimal findings on chest auscultation, and small segmental infiltrates on chest x-ray. In elderly patients, pneumonia due to C. pneumoniae can be especially severe and may necessitate hospitalization and respiratory support.

Chronic infection with C. pneumoniae has been reported among patients with chronic obstructive pulmonary disease and may also play a role in the natural history of asthma, including exacerbations. The clinical symptoms of respiratory infections caused by C. pneumoniae are nonspecific and do not differ from those caused by other agents of atypical pneumonia, such as Mycoplasma pneumoniae.

DIAGNOSIS

Serology, PCR amplification, and culture can be used to diagnose C. pneumoniae infection. Serology has been the traditional method of diagnosing infection by C. pneumoniae. The gold standard serologic test is the MIF test (see section on diagnosis of C. trachomatis genital infection, above). Any antibody titer above 1:16 is considered significant evidence of exposure to chlamydiae. According to a CDC-sponsored expert working group, the diagnosis of acute C. pneumoniae infection requires demonstration of a fourfold rise in titer in paired serum samples. There are no official recommendations for diagnosis of chronic infections, although many research studies have used high titers of IgA as an indicator. The older CF tests and EIAs for LPS are not recommended, as they are not specific for C. pneumoniae but identify the chlamydiae only to the genus level. The organism is very difficult to grow in tissue culture but has been cultivated in HeLa cells, HEp-2 cells, and HL cells. Although NAATs are commercially available for C. trachomatis, only research-based PCR assays are available for C. pneumoniae.

TREATMENT

Acknowledgment

The authors wish to acknowledge the late Walter E. Stamm, MD, for his significant contributions to the field of Chlamydia research. Dr. Stamm wrote the chapters on chlamydiae for previous editions of Harrison’s Principles of Internal Medicine, and we thank the editors for permission to reproduce Figs. 85-1 and 85-2 as well as Table 85-1 from his chapter in the 17th edition. Dr. Stamm died on December 14, 2009, and this chapter is dedicated to him.