Epidemiology, clinical presentation, and diagnosis of Pneumocystis pulmonary infection in patients with HIV

INTRODUCTION — The incidence of Pneumocystis jirovecii (previously named Pneumocystis carinii) pneumonia has dramatically declined due to effective antiretroviral therapy (ART) and, to a lesser extent, the use of prophylaxis. Despite this decrease, it remains one of the leading causes of opportunistic infections among persons with human immunodeficiency virus (HIV) and low CD4 cell counts, such as those who are unaware of their HIV diagnoses or are not receiving medical care.

The general features of Pneumocystis pulmonary infection in the patient with HIV, including clinical presentation and diagnosis, will be reviewed here. An overview of extrapulmonary disease due to Pneumocystis infection will also be discussed. Treatment and prophylaxis of Pneumocystis infection in individuals with and without HIV are discussed elsewhere. (See "Treatment and prevention of Pneumocystis infection in patients with HIV" and "Treatment and prevention of Pneumocystis pneumonia in patients without HIV".)

MICROBIOLOGY AND TERMINOLOGY — Pneumocystis is currently recognized as a fungus based upon ribosomal ribonucleic acid (RNA) and other gene sequence homologies, the composition of the cell wall, and the structure of key enzymes [1]. Prior to being identified as a fungus, the taxonomic classification of Pneumocystis as a genus of protozoan organisms had been questioned for several years. However, Pneumocystis organisms are atypical fungi as they do not grow in fungal culture, they respond to some antiparasitic agents, and they have a cell wall that contains cholesterol rather than ergosterol [2]. The life cycle consists of the trophic form, a precystic form, and the cystic form [3]. The trophic form predominates over the cystic form during infection.

The nomenclature for Pneumocystis has also changed; the species that infects rats is called P. carinii; and the one that infects humans, P. jirovecii [4]. P. jirovecii is now designated as the species name to use in publications and references to human infections [4,5]. Therefore, we will use the abbreviation "PJP" to refer to P. jirovecii pneumonia throughout this topic, although some sources still use the abbreviation "PCP" to refer to the clinical entity of "Pneumocystis pneumonia."

PATHOGENESIS

●Transmission – The primary mode of transmission of P. jirovecii is via the airborne route. Serologic studies show that primary infection occurs early in life, with 75 percent of humans infected by the age of four years [6]. It was initially believed that PJP remained in a latent state unless the patient became immunosuppressed; however, this may not account for all cases of PJP. Animal and human studies have shown clearance of the organism, and there is increasing evidence of transmission from person to person and possibly through environmental reservoirs [7-11]. The role of colonization in humans may also be of importance to Pneumocystis transmission.

●Pathogen-host interaction – Pneumocystis exists almost exclusively within the alveoli of the lung [2]. The trophic forms first attach to the epithelium. Interaction of Pneumocystis with alveolar epithelial cells and alveolar macrophages initiates a cascade of cellular responses in both the organism and lung cells. Attachment of Pneumocystis to lung epithelial cells enhances Pneumocystis proliferation.

Alveolar macrophages are the primary resident phagocytes that mediate the clearance of the organisms from the lung. Phagocytosis, respiratory burst, and inflammatory activation of alveolar macrophages in response to Pneumocystis are impaired in persons with HIV and may contribute to the pathogenesis of infection. Accumulating evidence indicates that beta-glucan molecules, which are abundant in the cell wall of Pneumocystis, are important components that drive the initiation of the inflammatory response during PJP. The most abundant surface protein of Pneumocystis is the major surface glycoprotein (MSG) [12]. Variation of the expressed MSG may facilitate evasion of host immune responses.

●Host immunity – Immune control of Pneumocystis involves production of chemokines and inflammatory cytokines by alveolar macrophages and epithelial cells. In healthy individuals, CD4+ T cells coordinate the host inflammatory response by recruiting and activating additional immune effector cells including monocytes and macrophages, which are responsible for elimination of the organism. In contrast to CD4+ T cells, the role of CD8+ T cells in host defense against Pneumocystis is more controversial. However, CD8+ T cells may have some beneficial effect, particularly in a situation of chronic CD4+ depletion. In mice, the T-cell mediated inflammatory response can cause impaired lung compliance and gas exchange; as such, both CD4+ and CD8+ T cells may result in deleterious lung inflammation. In addition to T cells, neutrophils, recruited by CXCL2 and IL-8, also participate in lung inflammation during PJP.

●Colonization – Healthy individuals, as well as those with HIV, underlying lung disease, and/or immunosuppression, may harbor Pneumocystis in their respiratory tract despite being without any signs or symptoms of disease. These colonized individuals have generally been identified using polymerase chain reaction (PCR) assays. (See 'Choosing the diagnostic test' below.)

The prevalence of Pneumocystis colonization among healthy adults ranges from 0 to 20 percent [2], and it has also been demonstrated to be a common colonizer in hospitalized patients with bacterial and viral pneumonia [13].

The clinical significance of colonization is not well understood, but may be important for several reasons [14]:

•Colonized individuals may be at risk of developing pneumonia or transmitting infection.

•Colonized individuals receiving Pneumocystis prophylaxis may be at risk for developing drug resistance mutations.

•Ongoing colonization may trigger inflammation and local alveolar damage leading to lung diseases, such as chronic obstructive pulmonary disease.

EPIDEMIOLOGY

●Incidence – The incidence of PJP has dramatically decreased after the administration of effective antiretroviral therapy (ART) and the widespread use of PJP prophylaxis [10,15-19]. As an example, in a United States cohort of patients with HIV, there was a decrease in PJP incidence from 3.0 per 100 person years between 1994 to 1997 to 0.39 per 100 person years between 2008 to 2010 [18,20]. In Uganda, the frequency of PJP among patients with HIV hospitalized with suspected pneumonia who had negative sputum acid-fast bacilli smears and underwent bronchoscopy decreased from nearly 40 percent of bronchoscopies in 1999 to 2000 to less than ten percent in 2007 to 2008 [21,22]. In Spain, there was a decrease in PJP incidence from 1.3 per 100 person years in 2000 to 0.33 per 100 person years in 2013 [23].

The decrease in incidence with ART is due to both immunologic improvement as measured by CD4 cell count increases and the suppression of HIV RNA. For example, in a European cohort of individuals with HIV, the incidence of primary PJP was zero among those with a CD4 count between 100 to 200 cells/microL who were receiving ART and were virologically suppressed, independent of PJP prophylaxis [19].

Despite this decrease, PJP is still one of the leading causes of opportunistic infection in individuals with HIV [18,20]. Most cases occur in patients who are undiagnosed or are not receiving care [24,25].

●Risk factors – The main risk factor for PJP is advanced immunosuppression in patients not taking antiretroviral therapy. Other risk factors include a CD4 cell count less than 200 cells/microL, a CD4 cell percentage of less than 14 percent, previous episodes of PJP, oral thrush, recurrent bacterial pneumonia, unintentional weight loss, and higher plasma HIV RNA levels [26].

CLINICAL FEATURES OF PULMONARY DISEASE

Clinical manifestations — The clinical manifestations of PJP are most commonly gradual in onset, and are characterized by fever (80 to 100 percent), cough (95 percent), and dyspnea (95 percent) progressing over days to weeks [27]. The average patient has pulmonary symptoms for about three weeks before presentation. Patients may describe fatigue with usual activities (climbing stairs, speaking on the telephone, shaving) that previously were done without difficulty. The cough is generally nonproductive. Other symptoms include fatigue, chills, chest pain, and weight loss. Approximately 5 to 10 percent of patients are asymptomatic.

The most common findings on physical examination are fever (over 80 percent of patients have a temperature exceeding 38.1ºC) and tachypnea (60 percent). The most common adventitial sounds are crackles and rhonchi, but a normal chest examination occurs in 50 percent of cases. Oral thrush is a common co-infection.

Findings that are uncommon for PJP include presence of hemoptysis and/or pleural effusion.

Laboratory findings — There are several laboratory findings that are observed in patients with HIV and PJP.

●Low CD4 counts – The incidence of PJP in patients with HIV increases as the CD4 count decreases [15,28,29], with most cases occurring when the CD4 count drops below 200 cells/microL [28,30]. A CD4 cell count percentage of less than 14 percent is also commonly observed.

●Poor oxygenation – Hypoxia occurs with progression of PJP. The alveolar-arterial oxygen gradient is widened in more than 90 percent of patients, ranging from mild (alveolar-arterial O₂ difference <35 mmHg) to severe (alveolar-arterial O₂ difference >45 mmHg) (calculator 1) [31]. Oxygen desaturation can occur with exercise and is highly characteristic of PJP [32-34]. (See "Evaluation of pulmonary symptoms in persons with HIV", section on 'Pulse oximetry or arterial blood gas analysis'.)

●Abnormal diffusion capacity – Although diffusion capacity is difficult to obtain in some clinical evaluation settings, when available, it can be helpful in differentiating between PJP and other causes of pulmonary symptoms. PJP is highly unlikely if the diffusion lung capacity for carbon monoxide (DLCO) is normal (eg, 70 percent of the predicted value or greater). One prospective study of 306 patients with HIV and 467 episodes of worsening respiratory symptoms found that PJP pneumonia was present in less than 2 percent of patients with a normal or unchanged chest radiograph and a single breath DLCO >75 percent of the predicted value [35].

●Elevated lactate dehydrogenase level – In studies done before the availability of effective antiretroviral therapy (ART), an elevated lactate dehydrogenase (LDH) level was present in 90 percent of patients with HIV and PJP and had some prognostic significance. In one study, the mean LDH of PJP survivors was 340 IU, while the mean level of non-survivors was 447 IU [36]. More importantly, a rising LDH level despite appropriate treatment portends a poor prognosis [36,37].

●Elevated 1-3-beta-D-glucan – Elevated plasma levels of 1-3-beta-D-glucan, a component of the cell wall of P. jirovecii, have been found in patients with HIV and PJP. In a study of 282 patients with HIV, those with a diagnosis of PJP had significantly higher median beta-D-glucan levels than patients without the disease (408 versus 37 pg/mL) [38]. The use of this assay in supporting the diagnosis of PJP is discussed elsewhere. (See 'Choosing the diagnostic test' below and 'Establishing the diagnosis' below.)

Radiographic manifestations

●Chest radiographs – Chest radiographs are initially normal in up to one-fourth of patients with PJP. The most common radiographic abnormalities are diffuse, interstitial or alveolar infiltrates [39] (image 1). Upper lobe infiltrates and pneumothoraces can also be seen; however, a higher incidence of both of these findings can be seen in patients using aerosolized pentamidine prophylaxis [40-42]. Other less common radiographic presentations include lobar or segmental infiltrates, cysts, nodules, and/or pleural effusions [27,41,43].

●High resolution computed tomography – On high resolution computed tomography (HRCT), PJP pneumonia typically manifests as bilateral patchy or nodular ground glass opacities. HRCT has a high sensitivity for PJP among patients with HIV [44-46]. One study, for example, evaluated 51 patients with suspected PJP and normal, equivocal, or nonspecific chest radiograph findings; HRCT had a sensitivity of 100 percent and a specificity of 89 percent when the presence of typical HRCT findings was used to indicate possible PJP [45]. While such findings are suggestive, they do not provide definitive evidence of infection. However, a negative high resolution computed tomography scan makes the diagnosis of PJP highly unlikely.

●F¹⁸-fluoro-deoxyglucose positron emission tomography (FDG-PET) – Although FDG-PET is not generally necessary in the diagnosis of PJP, it may be helpful in a patient with a normal chest radiograph in whom a HRCT cannot be obtained or when the HRCT chest findings are subtle and other alternative diagnoses are suspected. In patients with PJP, PET imaging often demonstrates increased FDG uptake throughout both lungs, especially in the perihilar region. FDG-PET may be preferred to gallium-67 citrate scanning due to faster imaging time and less radiation exposure for the patient [47].

●Gallium-67 citrate scanning – Gallium-67 citrate scanning is sometimes used to screen for PJP in suspected individuals with a normal chest radiograph but in whom a HRCT cannot be obtained. Nuclear scanning is a highly sensitive test in patients with PJP, demonstrating intense, diffuse bilateral uptake [48]. This test is rarely used for diagnosis today.

EVALUATION AND DIAGNOSIS — The diagnostic evaluation for PJP consists of identifying high-risk patients with a clinical presentation consistent with PJP, selecting the optimal respiratory specimen for testing, and choosing the test(s) to perform on the respiratory specimen. Each step is discussed in detail below.

The initial approach to the patient with HIV and pulmonary symptoms is discussed separately. (See "Evaluation of pulmonary symptoms in persons with HIV".)

When to suspect pneumocystis pneumonia — PJP should be strongly suspected in a patient with HIV not receiving antiretroviral therapy (ART) and a CD4 cell count less than 200 cells/microL who demonstrates symptoms/signs characteristic of the infection; particularly dyspnea, hypoxemia, cough, diffuse, and interstitial or alveolar infiltrates on chest radiograph or high-resolution computed tomography (HRCT).

Empiric treatment should be initiated in acutely ill patients in whom there is a high clinical suspicion for PJP, as obtaining appropriate specimens and processing of tests may take several days. Additionally, the initiation of empiric therapy does not preclude obtaining a definitive diagnosis since cysts and deoxyribonucleic acid (DNA) may persist for days to weeks after appropriate therapy has been administered [49,50]. (See "Treatment and prevention of Pneumocystis infection in patients with HIV", section on 'Treatment'.)

Choosing the optimal respiratory specimen — There are several respiratory and lung tissue specimens that are available for staining and polymerase chain reaction (PCR) testing. The most common specimens include induced sputum, bronchoalveolar lavage, and endotracheal aspirate. The type of specimen obtained depends on the respiratory status of the patient, concern for alternative pathogens, the institution where the specimen is being processed, and the risks of the procedure.

●Sputum induction – In most cases, sputum induction via the inhalation of hypertonic saline is the initial step in the attempt to obtain an adequate specimen. This is the least invasive method and should be sought in patients who are able to provide one (eg, patients with an appropriate mental status who are able to follow directions).

While the specificity of this method can approach 100 percent, the sensitivity is highly variable, ranging from 55 to approximately 90 percent for immunofluorescent staining (also known as direct fluorescent antibody [DFA] staining) [51-53] and 85 to 100 percent for PCR testing [54-56]. Factors that affect the accuracy of sputum induction include the type of testing (DFA versus PCR) that is used, the quality of the specimen, the burden of organisms (particularly in the setting of prophylactic therapy), and the expertise of the laboratory in interpreting the specimen.

For some patients, the induced sputum may not be diagnostic and negative immunofluorescent staining DFA and/or PCR results on an induced sputum specimen do not rule out PJP. The ability to detect Pneumocystis can be reduced in patients receiving prophylactic therapy, particularly those receiving aerosolized pentamidine [40,57].

For patients in whom an induced sputum is negative or cannot be collected, the bronchoalveolar lavage is the next optimal specimen provided sufficient uncertainty remains about the diagnosis, and empiric therapy is not deemed appropriate. (See 'Establishing the diagnosis' below.)

●Bronchoalveolar lavage and bronchoscopic sampling – In many cases, bronchoalveolar lavage (BAL) specimens are needed to make a definitive diagnosis of PJP. Bronchoalveolar lavage is the best option for patients who cannot produce an induced sputum or for whom induced sputum or endotracheal aspirate was non-diagnostic. Therefore, if sputum induction is negative or cannot be performed (eg, the patient cannot cooperate, is too dyspneic, or is unable to produce a specimen), fiberoptic bronchoscopy with BAL is the next recommended step.

BAL alone has a diagnostic yield of 90 to 100 percent in patients with HIV. To increase its sensitivity, site-directed lavage can be used in patients with focal infiltrates; this involves sampling the most heavily involved lobes on chest radiograph. One study has shown that the combination of site-directed lavage and DFA staining increased the sensitivity of BAL from 80 to 98 percent [58]. In another study of patients with suspected PJP and negative induced sputum testing with DFA, acquisition of a BAL sample led to a diagnosis in 51 percent of patients [59]. Testing with PCR may further add to the sensitivity of BAL.

Transbronchial biopsy, which has a diagnostic yield of up to 100 percent, can be added if the presentation for PJP is atypical and other diagnoses with good yield on transbronchial biopsy (eg, cytomegalovirus infection, sarcoidosis, hypersensitivity pneumonitis) are also being considered.

The major concerns with using BAL are the potential risks of the procedure. These include respiratory failure (rare) and fever (common). Transbronchial biopsy can be complicated by hemoptysis and pneumothorax (the latter occurring in less than 2 percent).

●Endotracheal aspirates in intubated patients – Endotracheal aspirates from intubated and mechanically ventilated patients have a high sensitivity for the detection of PJP. As an example, one study of 31 intubated patients found that endotracheal aspirates examined with immunostaining techniques had a sensitivity of 92 percent for PJP, with immunostained BAL specimens serving as the reference standard [60]. Thus, examination of endotracheal aspirates in intubated patients may obviate the need for bronchoscopy in many cases, as a negative DFA or PCR test on an endotracheal aspirate has a high negative predictive value.

●Other alternatives – If induced sputum, BAL, or endotracheal aspirate cannot be collected safely, we reassess the patient's risk for PJP (eg, CD4 count, whether the patient was taking ART or PJP prophylaxis), clinical presentation, and the availability of non-specific tests such as beta-D-glucan to see if a presumptive diagnosis can be made without resorting to more invasive specimen collection techniques. (See 'Establishing the diagnosis' below.)

Many patients with risk factors for PJP and characteristic clinical presentations can be treated empirically even with negative induced sputum and/or BAL testing without resorting to more invasive methods. This strategy is often supported by positive non-specific tests (eg, beta-D-glucan, lactate dehydrogenase [LDH]). (See 'Laboratory findings' above and 'Choosing the diagnostic test' below.)

However, in patients in whom an alternative diagnosis is likely, the use of these techniques may be necessary. Options include transthoracic needle biopsy or lung biopsy performed either via thoracotomy or by video-associated thoracoscopic surgery. However, the risks with these procedures, which are significant, must be weighed against the need for an accurate and definitive diagnosis.

•Transthoracic needle biopsy, which has a high diagnostic yield, is associated with a 30 percent incidence of pneumothorax.

•Lung biopsy, either by thoracotomy or by video-assisted thoracoscopic surgery, can be performed with a sensitivity of 95 to 100 percent for the diagnosis of PJP [26]. Histology on lung biopsy demonstrates the formation of a foamy, eosinophilic alveolar exudate; severe cases are associated with edema and interstitial fibrosis [61].

We do not use oral washes, nasopharyngeal aspirates, serum, or urine samples for the diagnosis of PJP due to the lack of consistent data supporting their use [62].

Choosing the diagnostic test — A collected respiratory specimen can be sent for general fungal staining, specific immunofluorescent Pneumocystis staining, and PCR. Since Pneumocystis cannot be cultured, making a definitive diagnosis relies upon the visualization of the cystic or trophic forms in appropriate specimens and/or identification of Pneumocystis deoxyribonucleic acid on PCR testing. (See 'Establishing the diagnosis' below.)

●Respiratory specimen tests

•PCR testing – PCR of respiratory fluid is increasingly used to make the diagnosis of PJP in both patients with and without HIV [31,63-65], and several clinical laboratories offer this test. PCR testing can help confirm the diagnosis in clinically suspect cases with negative sputum or BAL smears [66,67]. A disadvantage is that PCR cannot distinguish between colonization and disease, although studies have shown the likelihood of finding just colonization is low when the test is ordered in a population at risk of PJP [10]. As an example, in a retrospective study that compared rates of P. jirovecii detection before and after the introduction of PCR testing at a tertiary care center, Pneumocystis was detected in 11 of 1583 (0.69 percent) specimens by toluidine blue staining of microbiology, cytology, and/or pathology specimens in the year prior to the introduction of PCR testing, whereas 44 of 1457 (3 percent) of specimens were positive for Pneumocystis when PCR testing was used [68]. Of those with a positive PCR test, only three patients had low quantities of Pneumocystis DNA detected and were likely colonized; the remainder had evidence of true disease.

•Direct fluorescent antibody (DFA) staining – Immunofluorescent staining using fluorescein-labeled monoclonal antibodies (also known as direct fluorescent antibody) represents the preferred staining technique for the diagnosis of PJP and is more sensitive than the general stains [2,26,31]. Although DFA staining on a BAL specimen has a high sensitivity (greater than 90 percent) for PJP, negative DFA staining on an induced sputum or endotracheal aspirate specimen is not sensitive enough to exclude a diagnosis of PJP.

•General fungal staining – General fungal stains that selectively stain the cell wall of the cystic form include Gomori-methenamine silver (picture 1), cresyl violet, Gram-Weigert, and toluidine blue O. Wright-Giemsa and Diff-Quick detect both the cystic and trophic forms, but do not stain the cell wall (picture 2). Other agents that can be useful include the Papanicolaou stain and Calcofluor white. These fungal stains are widely available but the sensitivity to detect PJP is low [62].

●Beta-D-glucan assay – 1,3-beta-D-glucan is a cell wall component of many fungi, including Pneumocystis. Therefore, it is not specific for Pneumocystis infection and should be used as an adjunct to the diagnosis of PJP. If available with quick turn-around time, the beta-D-glucan assay is useful while awaiting the microscopy and/or PCR results of a respiratory fluid specimen or in situations in which respiratory sampling cannot be performed safely. (See 'Establishing the diagnosis' below.)

Support for the diagnostic utility of beta-D-glucan levels is provided by a study of 282 patients with HIV in which levels greater than 80 pg/mL were associated with a sensitivity and specificity of 92 and 65 percent for the diagnosis of PJP, respectively [38]. Further analysis showed that, among individuals with advanced immunosuppression and respiratory symptoms, the positive predictive value for PJP of a beta-glucan level of greater than 80 pg/mL was 96 and the negative predictive value of a negative beta-D-glucan level was 60 percent, respectively [69].

Potential confounding factors must be considered when interpreting the results of this test. Elevated levels can also be observed in patients infected with other fungi (in particular histoplasmosis), and false positives can be seen as a result of other clinical variables (eg, infections with certain bacteria, albumin infusion, use of cellulose filters/membranes for intravenous administrations and hemodialysis). False negative results are less common but can occur. In a meta-analysis of 23 observational studies the negative predictive value of a beta-D-glucan level <80 pg/mL was 95 percent when pre-test probability was intermediate (or around 50 percent) [70]. (See "Diagnosis of invasive aspergillosis", section on 'Beta-D-glucan assay'.)

Establishing the diagnosis — The approach to diagnosis includes definitive microbiologic identification of the organism via staining and/or PCR testing of a respiratory specimen or making a presumptive diagnosis based on the patient's risk, clinical presentation, and beta-D-glucan level. Obtaining a definitive diagnosis is preferred since treatment requires a prolonged course of possibly toxic therapy, and coinfections are common in patients with advanced immunodeficiency. (See 'Evaluation for coinfection' below.)

There are multiple testing strategies that can be used to make a definitive or presumptive diagnosis of PJP. The approach varies across institutions and practices based on the availability of each of these tests.

If all tests are available, we prefer to use PCR with the addition of beta-D-glucan in select cases when the clinical presentation is not straight forward or if there is concern for colonization. If PCR is not available, DFA is a reasonable alternative.

●If PCR is readily available – If PCR is available, the diagnosis of PJP can be made quickly and reliably with PCR alone based on a positive test result. Although a positive PCR result does not distinguish between colonization and true infection, the likelihood of colonization without infection in a patient with HIV, low CD4 count, and a clinical presentation consistent with PJP is unlikely.

If beta-D-glucan test is available, it can help distinguish colonization versus true infection in patients with atypical clinical presentations. A positive PCR and an elevated beta-D-glucan level is consistent with Pneumocystis infection, while a positive PCR with a low beta-D-glucan level may warrant workup for other potential causes of symptoms. A negative PCR test in the setting of an elevated beta-D-glucan level strongly supports a non-Pneumocystis cause of the high beta-D-glucan level.

●If PCR is not readily available – If PCR is not available, then visualization of the cystic or trophic forms in respiratory specimens by DFA or general fungal stains can also be used to make a definitive diagnosis of PJP. However, sensitivity of general fungal stains is low on all respiratory specimens and a negative fungal stain should not be used to rule out PJP. Similarly, if DFA is negative on an induced sputum or endotracheal aspirate specimen, the sensitivity of the test is not high enough to rule out PJP, and further respiratory specimens (eg, BAL) or diagnostic methods should be sought. (See 'Choosing the optimal respiratory specimen' above.)

If beta-D-glucan testing is available, a negative beta-D-glucan result can lower the post-test probability of the patient having PJP, especially in cases with an atypical clinical presentation for PJP or in those in whom alternative diagnosis are suspected.

●If neither PCR nor DFA is readily available or no respiratory specimen can be obtained – There are times when a definitive diagnosis cannot be made due to a low burden of organisms and/or the inability to obtain the necessary specimen. In these situations, a decision must be made whether to continue treatment. If neither of the microbiological tests are available or a respiratory specimen cannot be obtained, a presumptive diagnosis can still be made clinically based on the patient's clinical presentation, and if available, a positive beta-D-glucan. A positive beta-D-glucan result in the setting of a clinical and radiographic presentation consistent with PJP makes PJP very likely and a presumptive diagnosis can be established based on these criteria [26].

In patients without evidence of an alternative and/or concurrent diagnosis, we continue therapy for presumed PJP in individuals with all of the following clinical features:

•Advanced immunosuppression (CD4 cell count less than 200 cells/microL) in a patient not on ART.

•Clinical signs and symptoms such as cough, fever, dyspnea, hypoxemia (especially with exercise).

•Radiographic findings consistent with PJP on chest radiograph (interstitial, or alveolar infiltrates) or HRCT (patchy or nodular ground-glass attenuation).

•An elevated 1-3-beta-D-glucan level (defined as greater than 80 pg/mL) [69].

These patients must be closely monitored for failure to respond to treatment or clinical deterioration. In those cases, a more extensive work up may be indicated. (See 'Choosing the optimal respiratory specimen' above and 'Evaluation for coinfection' below and 'Differential diagnosis' below.)

Evaluation for coinfection — Approximately 15 percent of patients with PJP have an alternative or co-occurring infection/disease [26]; this rate may be higher in resource-limited settings where coinfection with Mycobacterium tuberculosis is more likely to occur [71]. The differential during evaluation and workup of symptoms should remain broad even in patients with a classic clinical presentation of PJP. Additionally, even after a definitive diagnosis of PJP is established, close monitoring during therapy to confirm the patient is improving is necessary to make sure no other co-infections are present. (See 'Differential diagnosis' below.)

DIFFERENTIAL DIAGNOSIS — Patients with HIV may have symptoms and/or signs that mimic PJP which are due to a wide variety of disease processes. Considerations include acute bronchitis, pneumonia due to bacteria, fungi, viruses and mycobacteria, neoplasm, drug hypersensitivity, pulmonary hypertension, and cardiomyopathy. These diseases may present with atypical signs and/or symptoms in patients with advanced immunosuppression and the clinician must be diligent in ensuring that they are not responsible for the underlying clinical presentation. This is discussed separately. (See "Evaluation of pulmonary symptoms in persons with HIV".)

Pulmonary infections with specific organisms are a significant concern in patients with HIV and CD4 counts less than 200 cells/microL. These include tuberculosis, nontuberculous mycobacteria, several different fungi, toxoplasma, cytomegalovirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and influenza. Kaposi sarcoma (KS) involving the lung is also a concern, particularly in patients with CD4 counts less than 100 cells/microL.

●Tuberculosis – Patients with tuberculosis (TB) present with fever, cough, weight loss, night sweats and malaise. As immunity declines, the frequency of pulmonary cavitation, which is the hallmark of pulmonary TB in adults, becomes progressively less common [72,73]. In individuals with HIV and advanced immunosuppression, the findings on chest radiographs can vary, ranging from no evidence of disease to a miliary pattern. (See "Diagnosis of pulmonary tuberculosis in adults".)

Compared with PJP, tuberculosis is associated with more severe constitutional symptoms. The incidence of TB in individuals with HIV varies markedly depending on epidemiology. Most cases in the United States now occur in individuals originally from countries where TB is highly endemic, or in those with risk factors for exposure (eg, prisoners, injection drug users, household contacts of active TB cases).

●Nontuberculous mycobacteria – There are a variety of nontuberculous mycobacteria (NTM) that can cause disease in patients with HIV, especially in those with CD4 cell counts less than 50 cells/microL. Most disease in HIV secondary to NTM presents as disseminated disease, without respiratory symptoms or pulmonary involvement; this is particularly the case for Mycobacterium avium complex (MAC). This manifestation of MAC is different from the multifocal nodular disease (with "tree in bud" radiographic opacities) seen in immunocompetent hosts, such as those with bronchiectasis. However, on rare occasions, disease due to M. kansasii and M. xenopi can present with localized pulmonary findings.

●Fungi – Patients with HIV who are from regions known to have endemic fungal infections can present with disseminated diseases that may mimic PJP. The most important of these is disseminated histoplasmosis. Both diagnoses should be considered in a patient with fever, cough, and diffuse interstitial infiltrates who is from a histoplasmosis-endemic area. Elevated beta-glucan levels are observed in both PJP and histoplasmosis. Findings suggestive of histoplasmosis include adenopathy, hepatosplenomegaly, and/or the presence of oral or other mucosal ulcerations. The diagnosis is confirmed with histoplasmosis antigen testing. Other fungi, including cryptococcus and coccidioides, can also mimic PJP. (See "Epidemiology, clinical manifestations, and diagnosis of histoplasmosis in patients with HIV" and "Cryptococcus neoformans infection outside the central nervous system", section on 'HIV-positive patients'.)

●Toxoplasmosis – Pneumonitis, as an extracerebral manifestation of toxoplasmosis, presents with fever, dyspnea, and non-productive cough [74]. Chest radiographs typically have reticulonodular infiltrates. The clinical picture may therefore be indistinguishable from PJP. Toxoplasma, which is a much less common respiratory pathogen than Pneumocystis in patients with HIV and low CD4 counts, can be identified in bronchoalveolar lavage (BAL) fluid [74,75]. (See "Toxoplasmosis in patients with HIV", section on 'Pneumonitis'.)

●Cytomegalovirus – Pneumonitis due to Cytomegalovirus (CMV) typically occurs in patients with HIV and CD4 cell counts less than 50 cells/microL. CMV pneumonitis and PJP have similar clinical presentations, but CMV pneumonitis is much less common in patients with HIV. A definitive diagnosis of CMV pneumonitis requires observing CMV inclusion bodies on biopsy.

●Influenza – Individuals with HIV are considered at high risk for complications of influenza, one of which is primary influenza pneumonia. This presents with the acute onset of a severe viral syndrome (fever, myalgias, headache) followed by progressive respiratory symptoms such as dyspnea and possibly cyanosis. Typical radiographic manifestations include bilateral reticular or reticulonodular opacities with or without superimposed consolidation. High-resolution computed tomography (CT) may show multifocal peribronchovascular or subpleural consolidation and/or ground glass opacities. Unlike the acute onset of influenza pneumonia, symptoms associated with PJP occur in a subacute fashion. (See "Seasonal influenza in adults: Clinical manifestations and diagnosis", section on 'Pneumonia'.)

●COVID-19 – Patients with coronavirus disease 2019 (COVID-19) and Pneumocystis pneumonia can present with dry cough and oxygen desaturation with ambulation. In addition, both can have ground-glass opacities on chest CT. Given these similarities, the diagnosis of Pneumocystis may not be considered, especially when COVID-19 incidence is high [76]. Although chest CT findings are more likely to involve the upper lobes in patients with Pneumocystis pneumonia, whereas chest CT abnormalities often have a peripheral distribution and involve the lower lobes in patients with COVID-19, there are no pathognomonic radiographic findings that would lead to the exclusion of either diagnosis without further testing. In some patients, concurrent COVID-19 and Pneumocystis pneumonia have been reported [77]. The diagnosis of COVID-19 is typically based on the results of polymerase chain reaction (PCR) testing from a nasopharyngeal swab. (See "COVID-19: Clinical features" and "COVID-19: Diagnosis".)

●Kaposi sarcoma – KS may cause a multifocal nodular disease in individuals with HIV and CD4 counts less than 100 cells/microL. Although most patients with pulmonary symptoms have skin findings, up to 20 percent have no evidence of cutaneous disease. Direct visualization of characteristic lesions on bronchoscopy remains the gold standard for diagnosis; if bronchoscopy cannot be performed, findings with nuclear scans can help differentiate KS from PJP. (See "Pulmonary involvement in AIDS-related Kaposi sarcoma".)

EXTRAPULMONARY DISEASE — Extrapulmonary manifestations of Pneumocystis infection are very rare and have been observed in patients with very advanced HIV, as well as in those receiving second-line therapy for PJP prevention, including dapsone and aerosolized pentamidine. Current or prior pneumonia does not need to be present at the time of presentation, and disease can be restricted to a single site. However, when multiple noncontiguous sites are involved, pulmonary disease is often present [78].

Extrapulmonary Pneumocystis has been reported to involve the eye (typically the choroid layer), ear, thyroid, spleen, bone marrow, as well as multiple other sites. In most cases, the diagnosis is made by detection of cysts in GMS-stained, formalin fixed tissue. In some situations, such as when disease is localized to the eye, the diagnosis is based on the characteristic appearance of the lesion on examination (eg, multiple, focal, circumscribed, creamy to yellow-white, round or oval, deep choroidal lesions without overlying or surrounding inflammation or involvement of other ocular structures) and supported by the appropriate response to therapy. The prognosis is better when disease is limited to the eye or ear compared with disseminated disease at noncontiguous sites [78].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Opportunistic infections in adults with HIV".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword[s] of interest.)

●Basics topic (see "Patient education: Pneumocystis pneumonia (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Microbiology – Pneumocystis jiroveci is currently recognized as a fungus based upon ribosomal RNA and other gene sequence homologies, the composition of their cell walls, and the structure of key enzymes. (See 'Microbiology and terminology' above.)

●Epidemiology – Although the incidence is decreasing, Pneumocystis pneumonia (PJP) remains one of the leading causes of opportunistic infection in individuals with HIV. Most cases occur in patients who are not receiving antiretroviral therapy (ART), either because they are newly diagnosed with HIV or not engaged in care. (See 'Epidemiology' above.)

●Risk factors – Risk factors include not receiving ART, a CD4 cell count less than 200 cells/microL, a CD4 cell percentage of less than 14 percent, previous episodes of PJP, oral thrush, recurrent bacterial pneumonia, unintentional weight loss, and higher plasma HIV RNA levels. (See 'Epidemiology' above.)

●Clinical and radiographic manifestations – The clinical manifestations, which are usually gradual in onset, are characterized by fever, cough, and dyspnea progressing over days to weeks. Radiographic characteristics include diffuse, interstitial or alveolar infiltrates on chest radiograph and/or patchy or nodular ground-glass attenuation on high resolution computed tomography. Chest radiographs can be initially normal in up to one-fourth of patients with PJP. (See 'Clinical manifestations' above and 'Radiographic manifestations' above.)

●Evaluation and diagnosis

•When to suspect PJP – PJP should be suspected in a patient with HIV not on ART and a CD4 cell count less than 200 cells/microL who demonstrates symptoms/signs characteristic of the infection. Empiric therapy should be initiated while diagnostic evaluation is ongoing. (See 'When to suspect pneumocystis pneumonia' above and "Treatment and prevention of Pneumocystis infection in patients with HIV", section on 'Treatment'.)

•Choosing the optimal respiratory specimen – In most cases, sputum induction is the initial step in the attempt to obtain an adequate specimen. In many cases, however, bronchoalveolar lavage (BAL) specimens are required to make a definitive diagnosis of PJP. For intubated patients, endotracheal aspirate is another less invasive alternative to BAL. (See 'Choosing the optimal respiratory specimen' above.)

•Choosing the diagnosis test – A collected respiratory specimen can be sent for general fungal staining, specific immunofluorescent Pneumocystis staining, and polymerase chain reaction (PCR). We prefer to send the PCR when available, because of its high sensitivity and specificity. Beta-D-glucan is a serum test that can be used adjunctively in uncertain cases to help make the diagnosis. (See 'Choosing the diagnostic test' above.)

•Establishing the diagnosis – A definitive diagnosis of PJP requires microbiologic identification of the organism via staining and/or PCR testing of a respiratory specimen. If a definitive diagnosis is unable to be made due to a low burden of organisms and/or the inability to obtain the necessary specimen, we consider a presumptive diagnosis of PJP based on the patient's risk, clinical presentation, and beta-D-glucan level. (See 'Establishing the diagnosis' above.)

●Differential diagnosis – The differential diagnosis is broad, and includes acute bronchitis, pneumonia due to bacteria, fungi, viruses and mycobacteria, neoplasm, drug hypersensitivity, pulmonary hypertension, and cardiomyopathy. (See 'Differential diagnosis' above.)

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Patricia Tietjen, MD, who contributed to an earlier version of this topic review.

UpToDate also gratefully acknowledges John G Bartlett, MD (deceased), who contributed on earlier versions of this topic and was a founding Editor-in-Chief for UpToDate in Infectious Diseases.