Chapter 197 Pelvic Inflammatory Disease
• Neisseria gonorrhoeae (GC) and Chlamydia trachomatis (CT) most common, followed by Ureaplasma urealyticum, Mycoplasma hominis, Streptococcus species, Escherichia coli, Haemophilus influenzae, Peptostreptococcus, and Peptococcus*
Pelvic inflammatory disease (PID) is a categorical name for a range of pelvic infections and inflammations. The CDC defines PID by the presence of abdominal and adnexal tenderness and cervical motion tenderness in the absence of another definable cause of the patient’s symptoms. The diagnosis does not require the presence of elevated WBC, erythrocyte sedimentation rate (ESR), or of fever.
PID causes women to make an estimated 2.5 million outpatient visits to health care providers each year. One fourth of these women will suffer serious long-term sequelae and all are at risk for recurrence.1
The social and physical costs of PID are also significant: the risk of ectopic pregnancy increases sixfold after a single episode of PID, and there is an estimated 13% risk of infertility after one infection and a 70% risk after three infections.2 For these reasons as well as costs, prevention and aggressive full-spectrum care of the acute phase are vital.
Several organisms (listed in Table 197-13) are known to be implicated in the etiology of PID, but GC and CT are the most common. Because of the complexities and inconsistencies involved in sampling and laboratory verification, it is difficult to make definitive statements about the causal agents in PID. Asymptomatic chlamydial infections are an important cause of PID. An estimated 1 million new infections with GC occur in the United States annually. More than 4 million infections occur annually due to CT. The true incidence of CT infection is probably higher because not all states require reporting. A higher proportion of PID has been ascribed to CT than to GC.4
Data from Karchmer AW. Sexually transmitted diseases. In Dale DC, Federman DD, eds. Scientific American medicine. New York: Scientific American, 1996:7:XXII-10-13.
The incidence of GC in the United States increased by a factor of 2.7 from 1960 to 1980.5 The infection has reached pandemic proportions in the United States, where it is estimated that for each reported case, four cases go unreported.6 Gonococci have been recovered from the urethra of 2.2% of sexually active servicemen who had minimal or no symptoms and from 40% of asymptomatic males who had contact with symptomatic women.7 In addition, an estimated 80% of the men and women exposed to Gonococcus will develop infections.8
For a seemingly delicate and fastidious species, Neisseria has impressive infective abilities, preferring human columnar and transitional epithelia. In less than 1 hour after intercourse, Gonococcus can establish itself on the urethral mucosa, where it successfully resists the flow of urine.9 Favored sites in the lower female genital tract are Bartholin’s and Skene’s glands, the urethra, and the endocervical canal.6 Direct spreading can occur from the endocervix across the endometrial surface to the tubal mucosa, or migration can occur through subendothelial vascular and lymphatic channels.2 Perhaps the most common method of spreading, however, is by vector: GC attached to spermatozoa are physically carried to the fallopian tubes.6 Primary pathogens can also enter the upper tract from retrograde menstruation or uterine contractions during intercourse.10
In the acute state, the gonococci and polymorphonuclear lymphocytes accumulate in the subepithelial connective tissue, resulting in patchy destruction of the overlying mucosa.9,11 The consequent thinning of the mucosal lining is thought to facilitate the penetration of GC into deeper tissue.9 It is probably for this reason that gonococci are reported to survive only a short time in the fallopian tubes.12 However, they may be not only surviving but thriving. The descent of the microbe beyond the surfaces being examined makes detection difficult.12
Concomitant infections are known to occur with GC.13 Some researchers have even proposed that GC’s primary role appears to be paving the way for secondary invaders from normal vaginal flora, allowing access to the upper tract.14 The associated infection will frequently be CT, but a superinfection can also be present, in which case one will find that anaerobic bacteria have colonized as well.13
In the United States it is estimated that 20% to 30% of PID cases are caused by CT.3,15 Further, one study found that acute chlamydial PID may be subclinical or silent in 66% to 75% of the cases.16 Laboratory diagnosis of chlamydial infection is difficult, which, combined with CT’s propensity to be asymptomatic, renders thorough assessment of the scope of these infections nearly impossible.
A 5-year study conducted in urban Sweden showed that, despite a decrease in the incidence of gonococcal PID, the total number of cases of acute PID were unchanged or even increased in the last year of the study.17 Another European study, conducted in 1977, found that 62% of women with acute salpingitis had elevated (titer ≥ 1:64) chlamydial IgG antibodies.18 However, most other studies completed to date show much lower percentages. These numbers will no doubt change as diagnostic technology improves and the clinical presentation of CT is more widely understood.
Anaerobes are the organisms most commonly isolated from the fallopian tubes or culs-de-sac of patients with PID.19 Anaerobic bacteria are probably not the chief causative agents but rather opportunists, establishing themselves in unsuccessfully defended tissues.
Anaerobic infections are commonly found in immunocompromised hosts and are generally of endogenous origin.20 The cervix and vagina of a normal healthy woman contain both anaerobic and aerobic bacteria.21 Anaerobic infections establish themselves more often in older patients and in women with a history of prior PID.5
M. hominis has not been demonstrated as a sole etiologic agent but rather seems to be a common contributor to the polymicrobial milieu that is often discovered in PID. One study found M. hominis in cervical cultures from 81% of women patients with GC and 64% of those without GC.23
There are serious physical consequences for women who have had PID. It has been estimated that, in the postinfection state, one of four women suffers from one or more sequelae such as abdominal pain, infertility, or ectopic pregnancy.5 Dyspareunia is a symptom that is often not investigated but which, when relevant questions are asked, is frequently found in the post-PID sufferer.
Death from salpingitis is rare and is generally due to rupture of the tubo-ovarian abscess with subsequent peritonitis. A mortality rate of 5.2% to 5.9% has been calculated for tubo-ovarian abscesses; before 1950, mortality was 80% to 100%.24,25 Better diagnostic understanding of this complication, treatment with antibiotics, and prompt surgical intervention have phenomenally improved both morbidity and mortality statistics.
The Fitz-Hugh–Curtis syndrome is a perihepatitis complicating the primary condition of PID. Characteristic violin-string adhesions attach the liver to the abdominal wall.20,26 These adhesions are due to local peritonitis involving the anterior liver surface and the adjacent abdominal wall.27 Historically, GC was thought to be the main contributor to this syndrome, but CT is now found more frequently.20,21
Infertility is a serious concern. Once a woman has had PID, she is at risk for additional attacks. This is in part because, after the fallopian tubes have been damaged by the infectious process, normal defense mechanisms are impaired. Reinfection has been found to be the most important cause of infertility after PID.28 One study comparing the rate of nonsurgical infertility in 1973 with that of 1976 noted a 45% increase. This translates to 122,000 infertile couples per year.29 This increased incidence is consistent with the concurrent epidemic of sexually transmitted disease (STD)–associated PID.
An alarming statistical analysis has shown that for each 1000 girls born in 1950, a total of 138 had one or more bouts of PID by age 30; 26 were infertile because of the PID; and 9 had surgery for ectopic pregnancy.5 Further, it has been postulated that by the year 2000 there will be one episode of salpingitis for each two women reaching reproductive age in 1970.21
Ectopic pregnancy is a severe sequelae, and any woman with a history of PID faces a sevenfold to tenfold increased risk.5 Ectopic pregnancies tripled in the United States from 1967 to 1977. In 1977 ruptured ectopic pregnancies accounted for 12% of maternal deaths and were the leading cause of maternal death in nonwhite women.29 In another study that followed women with PID for 9.5 years, 12.8% were infertile after one infection, 35.5% after two, and 75% after three infections. The same study showed that 1 in 24 women with PID will have an ectopic pregnancy.30
In addition to the obvious factor of sexual contact, the main risk factors are age, use or history of use of an intrauterine device (IUD), and previous history of PID. An earlier “sexual debut” puts a young woman in a high-risk group for PID, especially when there are multiple sex partners. The risk in sexually active 15-year-olds is 1 in 8, whereas in the average 24-year-old it is 1 in 80.5 One interesting hypothesis for these data is that the cervical mucus in the younger woman may be estrogen-dominated, creating an environment that is more accessible to pathogens.22 Women with multiple partners have a 4.6-fold greater risk than women in monogamous relationships.21
A woman faces an increased risk of PID if she uses an IUD. Oral contraceptive (OC) users are somewhat less likely to have GC; on the other hand, they are at increased risk for chlamydial invasion. Birth control remains a potent issue, with barrier methods being the techniques of choice because of their decreased PID risk. It should also be noted that one author included his own clinical observation that women in his practice who had vasectomized men as partners only seldom developed PID.31
The last risk factor to be considered is iatrogenic. This occurs when invasive procedures have introduced pathogens, or in some other way disturbed the tract flora and induced PID. Among these procedures are the following:
The route by which the pathogens gain access to the upper female tract has only recently been explored. Menstruation, sperm, and trichomonads have all been shown to be important in the transportation of pathogens into the salpinx.
Often the onset of menses corresponds with the onset of an episode of PID. Infections occurring around the menses tend to be GC rather than CT, a clinical curiosity that may ultimately shed light on the etiology. One hypothesis is that menstrual regurgitation assists the inflammatory response by carrying sloughed endometrial epithelium, which may have attached GC or intracellular CT. These organisms can then proliferate in the tubal epithelium or on peritoneal surfaces.10,32
Human sperm has proved to be an interesting and multifaceted variable in the precipitation of PID. Some of the research targets bacteriospermia as a cause of infertility in men, findings that are clearly relevant to PID. Increasingly, STD research is noting the incidence of asymptomatic male carriers.6,7,31,33–36 A large population study discovered that 66% to 75% of the men who tested positive for GC were asymptomatic.7
Designed to travel during intercourse, sperm also serve as effective vectors. Researchers took a look at this most basic interaction between a man and a woman by means of a laboratory experiment. They introduced organisms into capillary tubes containing cervical mucus, either alone or with added spermatozoa, and observed microbial motility. Cervical mucus had already been considered an effective mechanical and immunologic barrier between the abundant flora of the vagina and the upper tract, and the test results confirmed this idea. However, they also demonstrated that organisms attached to sperm could easily traverse the length of the mucus column. This may be particularly important during menses, because sperm migration has been observed through menstrual plasma but not during the luteal phase or through the cervical mucus of pregnancy.
Electron microscopy has produced amazing photographic evidence of organisms attached to sperm.6,37 The mechanism observed with piliated GC is that pili twist together with the tails of the spermatozoa in a rope-in-a-spider-web arrangement around the bacteria. Sperm have also been found intimately associated with cytomegalovirus, Toxoplasma, Ureaplasma urealyticum,32 and CT.38
Motile trichomonads serve as another transporter of PID. They can ascend from the vagina to the fallopian tubes, carrying additional invaders. In fact, it has been observed that trichomonads are never isolated from humans when heavy bacterial contamination is absent.32