Chapter 30: Neisseria

The genus Neisseria contains the two gram-negative cocci which are established human pathogens The genus also contains many commensal species, most of which are harmless inhabitants of the upper respiratory and alimentary tracts. The pathogenic species are Neisseria meningitidis (meningococcus), a major cause of meningitis and bacteremia, and Neisseria gonorrhoeae (gonococcus), the cause of gonorrhea.

Neisseria typically appear in pairs (diplococcic) with the opposing sides flattened, imparting a “kidney bean” appearance (Figure 30–1). They are nonmotile, non–spore-forming, and non–acid-fast. Their cell walls are typical of gram-negative bacteria, with a peptidoglycan layer and an outer membrane containing polysaccharides complexed with lipid and protein. The structural elements of N meningitidis and N gonorrhoeae are the same, except that the meningococcus has a polysaccharide capsule external to the cell wall.

Gonococci and meningococci require an aerobic atmosphere with added carbon dioxide and enriched medium for optimal growth. Gonococci grow more slowly and are more fastidious than meningococci, which can grow on routine blood agar. All Neisseria are oxidase-positive. Species are defined by growth characteristics and patterns of carbohydrate fermentation. Procedures are also available to distinguish N gonorrhoeae and N meningitidis from the other Neisseria by immunoassay and nucleic acid amplification (NAA).

Both pathogenic species possess pili and outer membrane proteins (OMPs), which vary in their function and antigenic composition. In the study of these meningococcal and gonococcal proteins, investigators have assigned names for molecules which appear to have similar functions in pathogenesis. Table 30–1 is an attempt to show similarities and differences. It should be understood that the assignment of the same name (eg, PorA) to a protein found in both species does not mean they are identical. It does suggest that they have similar structure and function.

The outer membrane of the two pathogenic Neisseria contains a lipopolysaccharide (LPS) variant which differs from that of most other gram-negative bacteria. The major difference is that the polysaccharide side chains are shorter lacking the variable O-antigen units of most other gram-negative bacteria. This short-chain neisserial polymer is called lipooligosaccharide (LOS). The lipid A and core oligosaccharide are structurally and functionally similar to the LPS of other gram-negative bacteria and LOS has the same endotoxic power of LPS. The pili, OMPs, and LOS are antigenic and have been used in typing schemes.

BACTERIOLOGY

Meningococci produce medium-sized smooth colonies on blood agar plates after overnight incubation. Carbon dioxide enhances growth, but is not required. Thirteen serogroups have been defined based on the antigenic specificity of their polysaccharide capsule. The most important disease-producing serogroups are A, B, C, W-135, and Y. In addition to the group polysaccharides, individual N meningitidis strains may contain distinct classes of pili and OMPs including porins and adherence proteins, some of which have structural and functional similarities to those found in gonococci. The function of other OMPs is unknown. The N meningitidis genome employs multiple genetic mechanisms to alter its antigenic profile including transformation followed by homologous recombination. When these changes involve the capsular polysaccharide, it is called capsule switching and could be a mechanism for emergence of serogroups not included in a vaccine.

MENINGOCOCCAL DISEASE

EPIDEMIOLOGY

The combination of rapidly progressive disease and obvious person-to-person spread has long made meningococcal disease one of the most feared of all infections. In fact, meningococci are found in the nasopharyngeal flora of 3% to 25% of healthy individuals. Transmission occurs by inhalation of aerosolized respiratory droplets. Close, prolonged contact such as occurs in families and closed populations promotes transmission. The estimated attack rate among family members residing with an index case is 1000 times higher than in the general population; this fact is evidence of the contagious nature of meningococcal infection. Other factors that foster transmission are contact with a virulent strain and host susceptibility (lack of protective antibody). Typical settings of larger outbreaks are schools, dormitories, and camps for military recruits. In these close living circumstances, N meningitidis spreads readily among newly exposed individuals, but disease develops only in those who lack group-specific antibody.

The incidence of invasive meningococcal infection varies widely depending on age, geographic locale, and serogroup. In the United States, attack rates vary between 0.5 and 1.5 cases per 100 000 population, but in some countries rates as high as 25 per 100 000 have been sustained for some time. Most disease occurs in children 6 months to 5 years old with a second peak at 18 to 25 years of age (Figure 30–4). Most cases are sporadic or in small family or closed-population (school, day care center) outbreaks. B, C, Y, and W-135 are the most common serogroups in developed countries. Serogroup A strains tend to emerge every 10 to 15 years in large epidemics, with attack rates as high as 1000 per 100 000. Since the second half of the 20th century, serogroup A pandemics have been largely confined to China, Russia, the Middle East, and Africa.

PATHOGENESIS

The meningococcus is an exclusively human parasite; it can either exist as an apparently harmless member of the resident microbiota or produce acute disease. For most individuals, the carrier state is associated with acquisition of protective antibodies, but for some, spread from the nasopharynx to produce bacteremia, endotoxemia, and meningitis takes place too quickly for immunity to develop. Meningococcal pili (type IV) protruding through the capsule are the primary mediators of initial attachment to surface proteins (CD46) on nonciliated cells in the nasopharyngeal epithelium. This is a prelude to invasion. In this process, the pili aggregate the bacteria into microcolonies which move as a unit (twitching motility) on the epithelial cell surface. These units bind to microvilli and enter these cells in membrane-bound vesicles. Once inside, meningococci quickly pass through the cytoplasm, exiting into the submucosa and eventually the bloodstream (Figure 30–2). In the process, they damage the ciliated cells, possibly by direct release of endotoxin.

Once meningococci gain access to the submucosa, their ability to produce disease is enhanced by factors that allow them to scavenge essential nutrients like iron and evade the host immune response. As with other encapsulated bacteria, the polysaccharide capsule enables meningococci to resist complement-mediated bactericidal activity by binding serum factor H to their surface (see Chapter 22). Meningococci also have surface proteins which bind this downregulator of C3b deposition. In addition, the LOS side chains are able to incorporate sialic acid, another factor H binder, from host substrates. Like the capsule of group B streptococci (see Chapter 25), the capsules of group B and C meningococci also include sialic acid.

The most serious manifestations of meningococcal disease are related to its spread to the bloodstream and, its namesake, the meninges. The exact mechanism of CNS invasion is unclear but is probably related to the level of the bacteremia. It occurs in the choroid plexus with its exceptionally high rate of blood flow. After CNS invasion, an intense subarachnoid space inflammatory response is induced by the release of cell wall peptidoglycan fragments, LOS, and possibly other virulence factors. This causes the release of inflammatory cytokines. A prominent feature of meningococcal disease with or without CNS invasion is systemic endotoxin activity (see Manifestations). When grown in culture, N meningitidis readily releases endotoxin-containing blebs of its outer membrane from the cell surface as shown in Figure 30–3. It is not known whether this occurs in vivo, but the model of the meningococcus as a hyperproducer of endotoxin certainly fits with its most serious disease manifestations.

IMMUNITY

Immunity to meningococcal infections is related to group-specific antipolysaccharide antibody, which is bactericidal and facilitates phagocytosis. The bactericidal activity is due to complement-mediated cell lysis via the classical complement pathway. Individuals with deficiencies in the terminal complement components have an enhanced risk for meningococcal disease but not for other polysaccharide capsule pathogens, such as Haemophilus influenzae type b.

Through the first 12 years of life the incidence of meningococcal meningitis is inversely proportional to the percentage of the population with bactericidal antibody (Figure 30–4). The peak incidence of disease occurs between 6 months and 2 years of age. This corresponds to the nadir in the prevalence of bactericidal antibody in the general population. This is the time gap between loss of maternal transplacental antibody and the appearance of naturally acquired antibody. By adult life, serum antibody to one or more meningococcal serogroups is usually present, but an immune deficit remains for the serogroups not encountered in the local community. Infections appear when populations carrying virulent strains mix (college, summer camp, military barracks) allowing susceptible individuals acquire strains of serogroups to which they have no immunologic experience.

Protective antibody is stimulated by infection and through the carrier state, which produces immunity within a few weeks. The natural immunization shown in Figure 30–4 may not require colonization with every serogroup or even with N meningitidis, because antibody may be produced in response to cross-reactive polysaccharides possessed by other Neisseria or even other genera. For example, Escherichia coli strains of a particular serotype (K1) have a polysaccharide capsule identical to that of the group B meningococcus. These E coli also have enhanced potential to produce meningitis in neonates.

Purified capsular polysaccharides are immunogenic, generating T-cell–independent immune responses. As with other polysaccharide immunogens, these responses are not strong, lack memory, and mature slowly. In particular they may not yet be mature in early childhood when the risk of meningococcal disease is greatest. The group B polysaccharide differs from that of the other groups in failing to stimulate bactericidal antibody at all. This is believed to be due to the similarity of its sialic acid polymer to human neural cell adhesion molecules. That is, it is recognized as self.

MENINGOCOCCAL DISEASE: CLINICAL ASPECTS

MANIFESTATIONS

The most common form of meningococcal infection is acute purulent meningitis, with clinical and laboratory features similar to those of meningitis from other causes. A prominent feature of meningococcal meningitis is the appearance of scattered skin petechiae, which may evolve into ecchymoses or a diffuse petechial rash (Figure 30–5). These cutaneous manifestations are signs of the disseminated intravascular coagulation (DIC) syndrome, which is part of the endotoxic shock brought on by meningococcal bacteremia (meningococcemia). Meningococcemia sometimes occurs without meningitis and may progress to fulminant DIC and shock with bilateral hemorrhagic destruction of the adrenal glands (Waterhouse-Friderichsen syndrome). However, the disease is not always fulminant, and some patients have only low-grade fever, arthritis, and skin lesions that develop slowly over a period of days to weeks. Meningococci are a rare cause of other infections such as pneumonia, but it is striking that localized infections are almost never recognized in advance of systemic disease.

DIAGNOSIS

Direct Gram smears of cerebrospinal fluid (CSF) in meningitis usually demonstrate the typical bean-shaped, gram-negative diplococci (Figure 30–1). Definitive diagnosis is by culture of CSF, blood, or skin lesions. Although N meningitidis is reputed to be somewhat fragile, it requires no special laboratory handling for isolation from presumptively sterile sites such as blood and CSF. Growth is good on blood or chocolate agar after 18 hours of incubation. Speciation is based on carbohydrate fermentation patterns or immunologic tests. Serogrouping may be performed by slide agglutination methods but has no immediate clinical importance.

TREATMENT

Penicillin has long been the treatment of choice for meningococcal infections because of its high antimeningococcal activity and good CSF penetration. Although resistance mediated by both β-lactamase and altered penicillin-binding proteins (PBPs) has been reported, it is still rare. Third-generation cephalosporins such as ceftriaxone and cefotaxime are also effective and are treatments of choice for acute meningitis until the meningococcal etiology is proven. In countries where penicillin resistance is significant, cephalosporins become the first-line treatment. For those with β-lactam hypersensitivity moxifloxacin and chloramphenicol are alternatives.

PREVENTION

Until the development and spread of sulfonamide resistance in the 1960s, chemoprophylaxis with these agents was the primary means of preventing spread of meningococcal infections. Rifampin, ceftriaxone, or ciprofloxacin are now the primary chemoprophylactic agents. In the absence of resistance, penicillin is still not effective for prophylaxis, probably due to inadequate penetration into the uninflamed nasopharyngeal mucosa. Selection of cases to receive prophylaxis is based on epidemiologic assessment. Risk is highest for siblings of the index case and declines with increasing age. The closeness and duration of contact with the index case are also important. For example, an infant sibling sharing a room with a person with meningococcal disease would be at the highest risk. Typically, family members are given prophylaxis, but other adults are not. Common-sense exceptions, such as playmates and healthcare workers with very close contact (eg, mouth-to-mouth resuscitation), are made at the discretion of the physician or epidemiologist. The presence or absence of nasopharyngeal carriage of N meningitidis plays no role in this decision because it does not accurately predict risk of disease.

The first purified polysaccharide vaccines for any bacterial infection were developed at the Walter Reed Army Institute of Research driven by the impact of sulfonamide resistance on recruit camp outbreaks of meningococcal meningitis. These vaccines were shown to stimulate group-specific antibody and to prevent disease in military and adult civilian populations. A vaccine containing A, C, Y, and W-135 polysaccharides was licensed in the United States, but proved poorly immunogenic for infants and children under 2 years of age. This was a huge disappointment because young children are the largest group at risk (Figure 30–4). We now know the reason. Purified polysaccharide vaccines only stimulate T-cell–independent responses and these become fully developed only after 2 years of age. As with pneumococcal and H influenzae polysaccharide vaccines, this problem was overcome by conjugating the polysaccharide to a protein carrier (diphtheria toxoid). This Meningococcal Conjugate Vaccine Quadravalent (MCV4) stimulates T-cell–dependent responses, which are both stronger and present at an earlier age. Its use now licensed from 9 months of age on and universally recommended beginning at age 11 with boosters at 16 years.

It is also recommended down to the age of 9 months for anyone at high risk for meningococcal disease (complement deficiency, asplenia, HIV infection). Hopefully, further experience and solution of the group B problem (see later) will push universal application of this protection down to infants and toddlers as is done with the highly successful H influenzae Hib vaccine (see Chapter 31).

The protein conjugate approach faces a unique difficulty with the meningococcus—the failure of the group B polysaccharide to be immunogenic at all. If this is due to its similarity to a human neural cell adhesion molecule, as suspected, it may not be overcome simply by protein conjugation. Group B causes up to one-third of all disease, so no vaccine that omits it is likely to be completely successful. For this reason, other approaches such as the use of OMPs (eg, PorA) or serum factor H binding proteins are being pursued. In an approach called reverse vaccinology, genetically engineered vaccines based on genes detected the DNA sequence of the entire group B meningococcal genome hold the promise of defining proteins that would immunize against group B and hopefully all serogroups of N meningitidis.

BACTERIOLOGY

Neisseria gonorrhoeae grows well only on chocolate agar and other specialized media enriched to ensure its growth. It requires carbon dioxide supplementation. Small, smooth, nonpigmented colonies appear after 18 to 24 hours and are well developed (2-4 mm) after 48 hours. Gonococci possess numerous pili (type IV) which are structurally similar to those of meningococci and extend beyond the outer membrane (Figure 30–6), (Table 30–1). In addition to intergonoccocal and epithelial cell adherence these fibers contract causing twitching motility of entire microcolonies. They also facilitate direct uptake of DNA (transformation) by gonococci. The gonococcal outer membrane is composed of phospholipids, LOS, and several distinct OMPs. The OMPs include porins (Por1BA and Por1BB) and adherence proteins known as Opa. Opa proteins are a set of at least 12 proteins that get their name from the opaque appearance they give to colonies as a result of adhesion between gonococcal cells. A variable number of the Opa proteins may be expressed at any one time. Certain Opas are associated with invasion of epithelial cells.

ANTIGENIC VARIATION

Neisseria gonorrhoeae and N meningitidis are among several microorganisms whose surface structures are known to change antigenically from generation to generation during growth of a single strain. The mechanisms involved have been more extensively studied in gonococci but appear to be similar in both species. The major gonococcal structures known to undergo antigenic variation are pili, Opa proteins, and LOS. The genetic mechanisms are discussed in the following discussion and illustrated in Figure 22–5.

Gonococcal pili are antigenically variable to an extraordinary extent. There are multiple genetic mechanisms, but the most important one appears to be recombinational exchange between the multiple pilin genes present in the chromosome of every strain. Some of these genes are complete and able to express pilin (pilE). Others are not, due to lack of an effective promoter and are thus silent (pilS). When recombination between expression and silent loci results in the donation of new sequences to an expression locus, the result can be expression of a pilin with changes in its amino acid composition and thus its antigenicity. The recombination could also involve exogenous DNA from another cell or strain, because gonococci naturally take up species-specific DNA by transformation. The process is complex, involving other genes that play a role in the assembly of pili and their functional characteristics, such as cellular adhesion. The numerous possible outcomes include no pilin subunits, pilin subunits unable to assemble, mature pili with altered functional characteristics, and fully functional pili with a new antigenic makeup.

The multiple gonococcal Opa proteins are each encoded by separate genes scattered around the genome. Various combinations of these genes may be either “on” or “off” at any one time. The switch is set during the transcription of each Opa gene for the next cell generation. As a result of a process called replicative slippage, the number of repeats of particular gene sequence can vary. When the time comes for translation, the number of repeats determines whether the gene will be in or out of frame to translate its Opa protein. If it is in frame, the gene is “on”; if not, the switch is “off.” Variation in gonococcal LOS has been observed in volunteer subjects challenged with intraurethral N gonorrhoeae, but the genetic mechanism is unknown.

These changes in the gonococcal surface are random events which may or may not have survival value depending on the circumstances. During the early stages of infection there could be positive selection for the expression of pili and Opas that mediate adherence. If the host has antibodies against one or more of these proteins they would be removed and the infecting population would shift to cells expressing pili or Opas to which there is no immunologic experience. An example of how these antigenic variants could be selected is shown in Figure 30–7. Taken together, these multifactorial, antigenic variations of the gonococcal surface may serve the dual purposes of escape from immune surveillance and timely provision of the ligands required to bind to human cell receptors.

GONORRHEA

EPIDEMIOLOGY

Gonorrhea is one of our greatest public health problems. The hundreds of thousands of cases reported in the United States each year are felt to represent less than 50% of the true number and the rates for adolescents are alarmingly high and increasing by 10% a year. The highest rates are in women between the ages of 15 and 19 years and in men between the ages of 20 and 24 years. No truly effective means of control is yet in sight. Our ability to stem the tide of changed sexual mores continues to be hampered by lack of an effective means to detect asymptomatic cases, resistance of N gonorrhoeae to antibiotics (see Treatment), and, to some extent, lack of appreciation of the importance of this disease. The latter is evidenced by failure of patients to seek medical care and reluctance to report cases to public health authorities due to privacy concerns. In the minds of too many, syphilis is dreaded and “unclean,” whereas gonorrhea is only “the clap” (“clap” is from the archaic French clapoir, “a rabbit warren”; later, “a brothel”).

Gonorrhea is acquired by genital contact with an infected person. The major reservoir for continued spread is the asymptomatic patient. Screening programs and case contact studies have shown that almost 50% of infected women are asymptomatic or at least do not have symptoms usually associated with venereal infection. Most men (95%) have acute symptoms with infection. Many who are not treated become asymptomatic but remain infectious. Asymptomatic male and female patients can remain infectious for months. The attack rates for those engaging in sexual intercourse with an infected person are estimated to be 20% (female to male) to over 50% (male to female). The organism may also be transmitted by oral–genital contact or by rectal intercourse. When all these factors operate in a sexually active population, it is easy to explain the high prevalence of gonorrhea. Although gonococci can survive for brief periods on toilet seats, nonsexual transmission is extremely rare. Virtually all gonococci isolated from children can be traced to sexual abuse by an infected adult.

PATHOGENESIS

Attachment and Invasion

Gonococci are not normal inhabitants of the respiratory or genital microbiota. When introduced onto a mucosal surface by sexual contact with an infected individual, adherence ligands such as pili and Opa proteins allow initial attachment of the bacteria to receptors (CD46, CD66, integrins) on nonciliated epithelial cells (Figure 30–2). Initial attachment by the pili, is mediated by the active force they generate in movement of their microcolonies across the cell surface (Figure 30–6). This is followed by a tighter attachment owing to Opa proteins. This close binding provides an opportunity for other OMPs (Por1BA) to trigger signaling cascades activating multiple enzymatic systems within the host cell. These reactions lead to induction of phagocytosis of the gonococci in a process involving microfilaments and microtubules of the invaded cell. The microvilli surround the bacteria and appear to draw them into the host cell in the same manner as meningococci. Thus, after initial attachment the gonococcus induces the host cell to actively take it inside (Figure 30–2). Once inside, the bacteria transcytose the cell and exit through the basal membrane to enter the submucosa.

Survival in the Submucosa

Once in the submucosa, the bacteria must survive and resist innate host defenses as well as adaptive immune responses acquired from a previous infection. As with meningococci, siderophores on the gonococcal surface enable the organisms to scavenge iron needed for growth from human iron transport proteins. Although gonococci lack the polysaccharide capsule of the meningococcus, they still have multiple mechanisms that protect them against serum complement and antibody. One of these is LOS sialyation in which the gonococcus is able to incorporate host sialic acid onto its own surface. This provides a mechanism for blocking surface C3b deposition by direct LOS/sialic acid binding of factor H or by facilitating its binding to surface porins. Reversal of this sialylation may have a different effect in early events in gonococcal infection. Sialidases present in cervico-vaginal secretions have been shown to facilitate access of gonococcal LOS to cervical receptors. That is, an enzymatic product of the bacterial microbiota (sialidase) enhances infectivity of gonococci at this site.

Even when phagocytes do encounter gonococci, surface factors such as pili and Opa proteins interfere with effective phagocytosis. The organisms are also able to defend against oxidative killing inside the phagocyte by upregulation of catalase production and an efficient antioxidant defense system. Taken together, these factors provide ample evidence that killing by neutrophils is sufficiently retarded to allow prolonged survival of gonococci in mucosal and submucosal locations.

Spread and Dissemination

In contrast to meningococci, N gonorrhoeae bacteria tend to remain localized to genital structures, causing inflammation and local injury, which no doubt facilitates their continued venereal transmission. Purulent exudates containing “sticky” clusters of gonococci held together by Opa proteins could be the primary infectious unit. Infection may spread to deeper structures by progressive extension to adjacent mucosal and glandular epithelial cells. These include the prostate and epididymis in men and the paracervical glands and fallopian tubes in women (Figure 30–8). Spread to the fallopian tubes is facilitated by pilus-mediated twitching motility, attachment to sperm, and finally to the microvilli of nonciliated fallopian tube cells. Injury to the fallopian epithelium is mediated by the local effect of outer membrane LOS. Gonococci are also known to turn over their peptidoglycan rapidly during growth, releasing peptidoglycan fragments which are toxic to the ciliated epithelium of the fallopian tube.

In a small proportion of infections, organisms reach the bloodstream to produce disseminated gonococcal infection (DGI). When this happens, the systemic findings have their own pattern (see Manifestations) and seldom take on the endotoxic shock picture of meningococcemia. Although differences have been noted between N gonorrhoeae strains that remain localized and those that produce DGI, their connection to pathogenesis is unknown. Both DGI and salpingitis tend to begin during or shortly after completion of menses. This may relate to changes in the cervical mucus and reflux into the fallopian tubes during menses.

Genetic Regulation of Virulence

Through all the stages of gonorrhea, gonococci are able to use a particularly rich variety of genetic mechanisms in deployment of the virulence factors previously described at the right time. Some are regulatory responses to environmental cues, such as iron in relation to iron-binding proteins, whereas others involve changes in the genome. Antigenic changes in both pili and Opa proteins have been demonstrated in human infection, including the isolation of antigenic variants from different sites in the same patient. These presumably take place by the recombinational and translational mechanisms described above (see Antigenic Variation) as the organisms replicate in the patient.

IMMUNITY

The apparent lack of immunity to gonococcal infection has long been frustrating. Among sexually active persons with multiple partners, repeated infections are the rule rather than the exception.

Both serum and secretory antibodies are generated during natural infection, but the levels are generally low, even after repeated infections. Another aspect is that even when antibodies are formed, antigenic variation defeats their effectiveness and allows the gonococcus to escape immune surveillance. Antigenic variation of pili, Opa proteins, and LOS is particularly likely to be important. Outbreaks have been traced to a single strain that demonstrated multiple pilin variations and Opa types in repeated isolates from the same individual or from sexual partners. In experimental models, passive administration of antibody directed against one pilin type has been followed by emergence of new pilin variants presumably through the sequence illustrated in Figure 30–7. It appears that although some immunity to gonococcal infection is present, its effectiveness is compromised by the ability of the organism to change key structures during the course of infection.

GONORRHEA: CLINICAL ASPECTS

MANIFESTATIONS

Genital Gonorrhea

The clinical spectrum of gonorrhea differs substantially in men and women (Figure 30–8). In men, the primary site of infection is the urethra. Symptoms begin 2 to 7 days after infection and consist primarily of purulent urethral discharge and dysuria. Although uncommon, local extension can lead to epididymitis or prostatitis. The endocervix is the primary site in women, in whom symptoms include increased vaginal discharge, urinary frequency, dysuria, abdominal pain, and menstrual abnormalities. As mentioned previously, symptoms may be mild or absent in either sex, particularly women.

Other Local Infections

Rectal gonorrhea occurs after rectal intercourse or, in women, after contamination with infected vaginal secretions. This condition is generally asymptomatic, but may cause tenesmus, discharge, and rectal bleeding. Pharyngeal gonorrhea is transmitted by oral–genital sex and, again, may be asymptomatic. Sore throat and cervical adenitis may occur. Infection of other structures near primary infection sites, such as Bartholin glands in women, may lead to abscess formation.

Inoculation of gonococci into the conjunctiva produces a severe, acute, purulent conjunctivitis. Although this infection may occur at any age, the most serious form is gonococcal ophthalmia neonatorum, a disease acquired during childbirth by a newborn from an infected mother. The disease was formerly a common cause of blindness, which is now prevented by the administration of prophylactic topical eye drops or ointment (silver nitrate, erythromycin, or tetracycline) at birth.

Pelvic Inflammatory Disease

The clinical syndrome of pelvic inflammatory disease (PID) develops in 10% to 20% of women with gonorrhea. The findings include fever, lower abdominal pain (usually bilateral), adnexal tenderness, and leukocytosis with or without signs of local infection. These features are caused by spread of organisms along the fallopian tubes to produce salpingitis and into the pelvic cavity to produce pelvic peritonitis and abscesses (Figure 30–9). PID is also known to develop when other genital pathogens ascend by the same route. These organisms include anaerobes and Chlamydia trachomatis, which may appear alone or mixed with gonococci. The most serious complications of PID are infertility and ectopic pregnancy secondary to scarring of the fallopian tubes.

Disseminated Gonococcal Infection

Any of the local forms of gonorrhea or their extensions such as PID may lead to bacteremia. In the bacteremic DGI phase, the primary features are fever, migratory polyarthralgia, and a petechial, maculopapular, or pustular rash. Some of these features may be immunologically mediated. Gonococci are infrequently isolated from the skin or joints at this stage despite their presence in the blood. The bacteremia may lead to metastatic infections such as endocarditis and meningitis, but the most common is purulent arthritis. The arthritis typically follows the bacteremia and involves large joints such as elbows and knees. Gonococci are readily cultured from the pus.

DIAGNOSIS

Gram Smear

The presence of multiple pairs of bean-shaped, gram-negative diplococci within a neutrophil is highly characteristic of gonorrhea when the smear is from a genital site (Figure 30–1). The direct Gram smear is more than 95% sensitive and specific in symptomatic men. Unfortunately, it is only 50% to 70% sensitive in women, and its specificity is complicated by the presence of other bacteria in the female genital flora that have similar morphology. Experience is required in reading smears, particularly in women. A positive Gram smear is generally accepted as diagnostic in men. It should not be used as the sole source for diagnosis in women or when the findings have social (divorce) or legal (rape, child abuse) implications.

Culture

Attention to detail is necessary for isolation of the gonococcus because it is a fragile organism that is often mixed with hardier members of the genital flora. Success requires proper selection of culture sites, protection of specimens from environmental exposure, culture on appropriate media, and definitive laboratory identification. In men, the best specimen is urethral exudate or urethral scrapings (obtained with a loop or special swab). In women, cervical swabs are preferred over urethral or vaginal specimens. The highest diagnostic yield in women is with the combination of a cervical and an anal canal culture; this is because some patients with rectal gonorrhea have negative cervical cultures. Rectal cultures in men and throat cultures are needed only when indicated by sexual practices.

Swabs may be streaked directly onto culture medium or promptly transmitted (in less than 4 hours) to the laboratory in a suitable transport medium. Laboratory requests must specify the suspicion of gonorrhea so that media that satisfy the nutritional requirements of the gonococcus and inhibit competing normal flora can be seeded. The selective medium (eg, Martin-Lewis agar) is an enriched selective chocolate agar with antibiotics. The exact formulation changes but includes agents active against gram-positive bacteria (vancomycin), gram-negative bacteria (colistin, trimethoprim), and fungi (nystatin, anisomycin) at concentrations that do not inhibit N gonorrhoeae.

Direct Detection

Much effort has been directed at developing immunoassay and nucleic acid amplification (NAA) methods that detect gonococci in genital and urine specimens without culture. Such methods have particular importance for screening populations in which culture is impractical. After a series of improvements NAA methods are now considered the diagnostic standard. NAA results are considered diagnostic from genital sites (including urine) but may need to be confirmed by culture from other sites. The cost/benefit ratio of NAA tests has been improved by combining them with Chlamydia detection (see Chapter 39), which targets the same clinical population.

Serology

Attempts to develop a serologic test for gonorrhea have not yet achieved the needed sensitivity and specificity. A test that would detect the disease in asymptomatic patients would be very useful in control of this disease.

TREATMENT

Penicillin, which once was active against all known gonococci at extremely low concentrations (less than 0.1 μg/mL), is no longer used due to the development of multiple mechanisms of resistance.

Third-generation cephalosporins resistant to the β-lactamases prevalent in gonococci are now the standard. In addition, it is now recommended that all patients treated for gonorrhea also be treated for Chlamydia infection. For gonorrhea, ceftriaxone is given in a single intramuscular injection. For Chlamydia, either azithromycin or doxycycline (both oral) is added. Resistance rates up to 25% have taken fluoroquinolones out of the picture.

PREVENTION

Condoms provide a high degree of protection against both infection with N gonorrhoeae and transmission to a sexual partner. Spermicides and other vaginal foams and douches are not reliable protection. The classic public health methods of case–contact tracing and treatment are important, but difficult because of the size of the infected population. The availability of a good serologic test would greatly aid control, as it has for syphilis. Although candidate immunogens continue to be studied, the development of a vaccine is a high but distant goal.