Pneumocystis jirovecii (carinii) Pneumonia Imaging 

Updated: Jun 29, 2022
  • Author: Ali Nawaz Khan, MBBS, FRCS, FRCP, FRCR; Chief Editor: Eugene C Lin, MD  more...
  • Print

Practice Essentials

Pneumocystis jirovecii pneumonia (also known as pneumocystis pneumonia, or PCP; formerly P carinii pneumonia) is caused by the ubiquitous unicellular eukaryote P jirovecii. This organism is a rare cause of infection in the general population, but it is a frequent cause of morbidity and mortality in persons who are immune compromised, especially those with acquired immunodeficiency syndrome (AIDS). [1, 2, 3, 4]  

Although PCP is a frequent manifestation of AIDS, the granulomatous form is uncommon. Physicians should strongly suspect PCP in HIV-positive patients with nodular lung lesions and must remain aware that these lesions, if immune in origin, might not respond to antimicrobial therapy. [5]

Although PCP was historically associated with HIV/AIDS, a shift in demographics has resulted in increasing incidence in patients with hematologic malignancies and transplants. A granulomatous response to PCP infection is uncommon and most commonly presents as multiple randomly distributed nodules on chest imaging. Granulomatous PCP presents with similar clinical manifestations as typical pneumocystis pneumonia but usually is not detected by bronchoalveolar lavage and may require biopsy for definitive diagnosis. For this reason, the radiologist may be the first provider to suggest this diagnosis and to guide management. [6]

PCP is classified as a fungal pneumonia but does not respond to antifungal therapy. The incidence of PCP has decreased as a result of highly active antiretroviral therapy (HAART). Infection with multiple pathogens concurrently has become less common since the introduction of potent antiretroviral agents and effective prophylactic agents. Given that patients with PCP with a diagnosis of AIDS could be concomitantly infected with multiple pathogens, rapid accurate diagnosis and treatment may have a positive effect on outcome. [7]

Patients who do not have AIDS but are immune compromised and at risk for P jiroveciipneumonia include those with hematologic malignancies [4] ; organ transplant recipients [8, 9] ; and patients receiving long-term steroid or cytotoxic therapy, including individuals with systemic vasculitis or another autoimmune deficiency. Other patients with immune deficiency disorders who are at particular risk for PCP include those with thymic dysplasia, severe combined immunodeficiency, or hypogammaglobulinemia. Severe malnutrition may predispose individuals to PCP.

A subset of patients present with atypical clinical and radiographic features, termed "chronic PCP." These patients have a prolonged clinical course over months or years, with persistent stable symptoms and radiographic abnormalities corresponding to pathologic findings of interstitial fibrosis, traction bronchiectasis, and honeycombing.

In the non-AIDS group, several underlying conditions and immune defects could lead to different PCP presentations. Compared to other underlying diseases, PCP related to chronic lymphocytic leukemia is closer to PCP related to AIDS presentation (long duration of symptoms before diagnosis, high level of dyspnea, and low oxygen saturation at diagnosis). PCP presentations may vary according to the underlying reason for immunosuppression. Response to treatment and adjuvant steroid therapy should be analyzed. [10]

PCP is a common pathology in HIV-infected but also in uninfected immunocompromised individuals. With the coronavirus disease 2019 (COVID-19) pandemic, chest imaging is often used as a complementary tool in patient evaluation. The imaging finding is similar to that of many pulmonary pathologies. Chest computed tomography (CT) scanning is the gold-standard imaging modality and shows a central and diffuse distribution and a ground-glass pattern with septal thickening with a "crazy paving pattern." Diagnosis is confirmed by polymerase chain reaction with bronchoalveolar lavage fluid. Other immunochemical tests can be performed but are less specific. Trimethoprim-sulfamethoxazole is both curative and preventive in individuals at risk. [11]

Imaging modalities

Chest radiographs should be included in the initial evaluation for PCP. Frequently, these are the only images required. High-resolution CT (HRCT) scanning and, occasionally, gallium-67 (67Ga) scanning are useful in symptomatic patients for whom chest radiographic findings are normal or equivocal. [12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]

The hallmark finding of PCP on HRCT scans is diffuse ground-glass opacity (GGO), which reflects accumulation of intra-alveolar fibrin, debris, and organisms. [22, 23] The term "ground-glass" refers to parenchymal opacification, which does not obscure the underlying pulmonary architecture. This feature is usually seen in a bilateral, symmetric, predominantly perihilar distribution and may be geographic or mosaic in appearance, with areas of normal lung adjacent to areas of affected lung.

Accurately differentiating pneumocystis from cytomegalovirus pneumonia is crucial for appropriate therapy selection for patients with AIDS. In a study undertaken to compare CT features of pneumocystis pneumonia versus cytomegalovirus pneumonia in patients with AIDS and to identify clinical hallmarks to accurately distinguish these pathologies, consolidation, halo signs, and nodules (all P< 0.05) were significantly more frequent in patients with cytomegalovirus pneumonia than in those with pneumocystis pneumonia. Small nodules (32.5% in cytomegalovirus pneumonia, 6.41% in pneumocystis pneumonia; P< 0.001) without perilymphatic distribution were particularly common in patients with cytomegalovirus pneumonia. Large nodules were not found in any patients with cytomegalovirus pneumonia. Ground-glass opacity, reticulation, and bronchial wall thickening (all P > 0.05) were common in both groups. Analysis of consolidation, nodules, and halo signs may contribute to the differential diagnosis of pneumocystis pneumonia or cytomegalovirus pneumonia. However, some CT features considered typical in one or another disease appear with similar frequency in both cohorts of patients with AIDS. Researchers have concluded that CT features are potentially useful for the differential diagnosis of pneumocystis pneumonia and cytomegalovirus pneumonia in patients with AIDS. [25]

Chest radiography retains a key role in the diagnosis of pneumonia in immune compromised individuals. [20] Chest radiography retains its position as the prime modality for diagnosis and exclusion of pneumonia and for follow-up imaging to check for resolution and to evaluate potential complications. CT scanning is more sensitive and, with certain infections, more specific. Magnetic resonance imaging (MRI) provides an option for monitoring progress, although it has not yet replaced chest radiography or CT scanning as the initial diagnostic procedure. [18]

Findings on chest radiography may be normal in 10-39% of patients with PCP. On CT and 67Ga scanning, the appearance of PCP is nonspecific.

Ultrasonography may be useful in evaluation of systemic PCP infection (hepatic/splenic and renal microabscesses), but this imaging modality is of no value for assessment of pulmonary disease.

(See the images below.)

This radiograph depicts a diffuse, fine, reticular This radiograph depicts a diffuse, fine, reticular opacification as a result of Pneumocystis jirovecii pneumonia.
This chest radiograph shows bilateral upper lobe p This chest radiograph shows bilateral upper lobe pneumatoceles after Pneumocystis jirovecii infection in a patient with acquired immunodeficiency syndrome.
A magnified view of lung apices from a patient wit A magnified view of lung apices from a patient with human immunodeficiency virus infection. This image shows redistribution of Pneumocystis jirovecii pneumonia to the upper lobes following aerosolized pentamidine prophylaxis.
High-resolution CT (HRCT) scan in a 32-year-old ma High-resolution CT (HRCT) scan in a 32-year-old man with HIV infection showing ground-glass appearance due to Pneumocystis jirovecii pneumonia.

 

 

Next:

Radiography

In patients with PCP, chest radiographs classically demonstrate bilateral, diffuse, often perihilar, fine, reticular interstitial opacification, which may appear somewhat granular. This opacification progresses to air-space consolidation over 3 to 4 days, which may be followed by coarse reticulation as infection resolves. [24, 26, 27, 28]

(See the images below.)

This radiograph depicts a diffuse, fine, reticular This radiograph depicts a diffuse, fine, reticular opacification as a result of Pneumocystis jirovecii pneumonia.
This radiograph depicts the typical bilateral air- This radiograph depicts the typical bilateral air-space consolidation of Pneumocystis jirovecii pneumonia in a patient with acquired immunodeficiency virus infection.
This chest radiograph shows residual interstitial This chest radiograph shows residual interstitial opacities in a patient with a history of Pneumocystis jirovecii pneumonia.
A posteroanterior chest radiograph from a patient A posteroanterior chest radiograph from a patient with human immunodeficiency virus infection. This image shows redistribution of Pneumocystis jirovecii pneumonia to the upper lobes following aerosolized pentamidine prophylaxis.
An anteroposterior radiograph from a 33-year-old m An anteroposterior radiograph from a 33-year-old male patient with human immunodeficiency virus and Pneumocystis jirovecii pneumonia. This image shows features of a right-sided tension pneumothorax.
Bilateral spontaneous pneumothoraces resulting fro Bilateral spontaneous pneumothoraces resulting from Pneumocystis jirovecii pneumonia in a man with HIV infection that was previously undiagnosed.

Findings on chest radiography may be normal (in 10-39% of patients), or radiographic changes may lag behind clinical symptoms.

Trends in radiographic manifestations of PCP are changing; features that previously were considered to be unusual are seen with increasing frequency. [26]

Atypical radiographic patterns are reported to occur in 5% of patients and include cystic lung disease, spontaneous pneumothorax, and isolated lobar or focal consolidation, particularly with an upper lobe predominance.

Pulmonary nodules, which may be cavitated, have been described, but they are rarely seen in PCP. These nodules have been shown histologically to represent granulomas and are usually encountered early in the course of HIV infection, when the patient is still capable of mounting a granulomatous response.

Miliary nodularity, bronchiectasis, endobronchial lesions, and mediastinal lymphadenopathy (18%), which may show calcification, have been reported. [29]

Pleural effusions and hilar lymphadenopathy are uncommon. Indeed, the presence of an effusion should prompt the search for a different pathogen.

Cysts

Cysts are visible on chest radiography in 10% of patients, although these entities are appreciated far more commonly on HRCT scans (33%). Findings of cysts or pneumatoceles are not infrequent in patients with PCP.

Cysts may occur in the acute or postinfective period and range in number, size, shape, and distribution. They are commonly multiple, with a predilection for the upper lobes, and may be related to an ongoing or previous PCP infection.

The etiology of these cysts is unclear, but several hypotheses have been proposed, including release of elastase from alveolar macrophages, which causes tissue necrosis and cavitation; vascular invasion with subsequent infarction; and cavitation obstruction of small airways, leading to a ball-valve effect.

Radiologic-pathologic correlation has shown persistent infection in some cyst walls.

Spontaneous pneumothorax

Spontaneous pneumothorax may be a feature of PCP infection, with a reported incidence of approximately 6%, rising to approximately 35% in patients with cysts. Development of a spontaneous pneumothorax has important implications for the treatment and prognosis of patients because this condition tends to be refractory to conventional tube drainage, frequently requiring pleurodesis or surgical intervention. In addition, spontaneous pneumothorax is associated with a significantly higher mortality rate, particularly among patients on ventilation. Pneumothoraces are frequently bilateral.

Resolution

Findings on chest radiography usually resolve within 2 to 4 weeks with successful treatment. This resolution may be accelerated by the use of steroids. Occasionally, radiographic findings remain abnormal and images show reticular opacities, interstitial fibrosis, or focal scarring and/or nodularity.

Degree of confidence

Despite the presence of overlapping radiographic features in PCP, findings on chest radiography are often of diagnostic value. Usually, chest radiography is the only imaging required; its overall accuracy for the diagnosis of PCP is approximately 75%.

Chest radiographic findings may be normal in 5-30% of patients with a diagnosis of PCP. The literature reports a false-negative rate of 35- 40% when chest radiography is used for diagnosis of PCP. Adult respiratory distress syndrome, pulmonary edema, other opportunistic lung infections, lymphoma, and Kaposi sarcoma may mimic PCP.

Previous
Next:

Computed Tomography

High-resolution computed tomography (HRCT) scanning is more sensitive than chest radiography for detection and exclusion of PCP; these results may be positive when chest radiographic findings are normal. [16, 20, 22, 23]

The hallmark finding of PCP on HRCT scans is ground-glass attenuation, which is seen in more than 90% of patients and represents an exudative alveolitis. The term "ground-glass" refers to parenchymal opacification, which does not obscure the underlying pulmonary architecture. This usually occurs in a bilateral, symmetric, predominantly perihilar distribution and may be geographic or mosaic in appearance (56%), with areas of normal lung adjacent to areas of affected lung.

(See the images below.)

High-resolution computed tomography scan obtained High-resolution computed tomography scan obtained through the upper lobes in the prone position in a patient with a history of Pneumocystis jirovecii pneumonia. This image shows parenchymal and subpleural cysts and patchy fibrosis that resulted from Pneumocystis jirovecii infection.
High-resolution CT (HRCT) scan in a 32-year-old ma High-resolution CT (HRCT) scan in a 32-year-old man with HIV infection showing ground-glass appearance due to Pneumocystis jirovecii pneumonia.

Thickening of interlobular septa (due to edema) and foci of consolidation may be associated. Septal thickening in the subacute stage is usually extensive and represents organizing inflammatory infiltrate.

Degree of confidence

In the proper clinical setting, ground-glass attenuation on HRCT scans in patients with AIDS is virtually diagnostic of PCP, with diagnostic accuracy of approximately 94%. Normal HRCT findings virtually exclude the possibility of PCP.

Ground-glass attenuation is highly suggestive of PCP, but cytomegalovirus (CMV) pneumonitis [21]  and lymphoid interstitial pneumonia can (albeit infrequently) give rise to a similar appearance. However, CMV pneumonitis is rare in patients with CD4 counts greater than 50 cells/mm3.

Although patients with PCP may present with parenchymal nodules, this feature is more common in CMV infection; thus, the combination of ground-glass attenuation and nodularity is more likely to be secondary to CMV infection.

Motion artifacts and low lung volumes due to reduced inspiratory effort may occasionally have a spurious ground-glass appearance.

Ground-glass opacification can also be seen in conditions such as pulmonary edema, pulmonary hemorrhage, drug toxicity, other infections, and hypersensitivity pneumonitis. Clinical correlation usually allows exclusion of most of these differential diagnoses.

Hilar lymphadenopathy may occur in patients with tuberculosis, Mycobacterium avium complex (MAC) infection, fungal infection, Kaposi sarcoma, and AIDS-related lymphoma, but this condition is rare in patients with PCP.

Previous
Next:

Nuclear Imaging

Gallium citrate (Ga) is useful in investigation of fever of unknown origin (FUO) because it is taken up by areas of inflammation, infection, and tumor. Ga accumulates in PCP; this can facilitate detection of PCP in asymptomatic patients with AIDS in the absence of abnormal plain radiographic findings.

The most common pattern of radionuclide uptake in patients with PCP is diffuse pulmonary uptake. [12, 13, 14, 28]  A negative heart with diffuse pulmonary uptake in a patient with AIDS is indicative of PCP. However, uptake varies in patients treated with aerosolized pentamidine and is observed only in areas of the lungs not reached by the drug. Patchier uptake is seen with recurrent PCP; however, gallium scanning is expensive, is poorly tolerated by patients, and requires delayed scans at 48 hours. In practice, this study is little used.

Indium-111 (111In)-labeled autologous leukocytes accumulate in PCP, but the overall performance of this test in immunosuppressed patients is poor compared to 67Ga studies.

Clearance of technetium-99m (99mTc) diethylenetriamine pentaacetic acid (DTPA) aerosol across the alveolar-capillary membrane is accelerated in patients with PCP. The shortened half-life for clearance of radionuclide activity has been shown to be more sensitive than with 67Ga imaging. After effective therapy, shortened clearance times rapidly return to normal. [30]

Tc-99m-labeled nonspecific polyclonal human immunoglobulin (HIG) has been used for evaluation of patients with AIDS. [13, 31] Sensitivity varies from 0 to 100% in PCP. Similar to 67Ga scanning, 99mTc-labeled nonspecific polyclonal HIG appears to be more sensitive than chest radiography. The pattern of activity is usually diffuse, but focal uptake has been described.

Fab fragment of an antibody labeled with 99mTc has been used to image the infection in patients with AIDS; this fragment recognizes PCP. In a small series, sensitivity of 85.7% and specificity of 86.7% were achieved. [32]

Degree of confidence

Ga-67 scans are extremely sensitive for PCP, with reported sensitivities of 87-100%; however, the specificity of 67Ga imaging may vary considerably and reportedly ranges from 20 to 100%. This variation depends in part on the clinical practice and referral patterns. Specificity can be increased when diffuse pulmonary uptake of greater intensity than is seen in the liver is included in the diagnostic criteria. Discordance between pulmonary 67Ga uptake and negative chest radiographic findings in patients with AIDS can be used to increase specificity in detection of PCP.

Overall performance with uptake of radiolabeled leukocytes is poor in PCP; this technique should be reserved for imaging when bacterial pneumonia and infection are suggested at other sites in patients with AIDS and in patients who do not have AIDS but are immune suppressed.

Tc-99m DTPA aerosol clearance times provide a simple and noninvasive technique for follow-up imaging in patients receiving treatment for PCP. Although abnormalities in the clearance of 99mTc DTPA aerosol have been reported with other pulmonary infections in patients with AIDS, clearance time greater than 4.5% per minute has been shown to be specific for PCP in these patients.

The sensitivity and specificity of 99mTc-labeled HIG are too variable to warrant use of this technique in patients with AIDS-related PCP. Further large-scale studies are required to justify its use.

Ga-67 accumulates in lymphoma and in other malignant processes associated with AIDS.

Accelerated clearance of 99mTc DTPA aerosol is not specific for patients with PCP; this process has been reported with other pneumonitides associated with AIDS.

Previous