Skip to main content
British Journal of Clinical Pharmacology logoLink to British Journal of Clinical Pharmacology
. 1999 Jun;47(6):605–609. doi: 10.1046/j.1365-2125.1999.00966.x

The management of Pneumocystis carinii pneumonia

F J Vilar 1, S H Khoo 1, T Walley 1
PMCID: PMC2014249  PMID: 10383536

Introduction

Although advances in the management of human immune-deficiency virus (HIV) have dramatically reduced the incidence of opportunistic infections, Pneumocystis carinii pneumonia (PCP) remains an important respiratory infection, particularly when it is the presenting feature of previously undiagnosed HIV infection. The incidence of PCP as an AIDS indicator disease amongst those previously known to have HIV has fallen in the United Kingdom from 35% in 1990 to 22% in 1997, but the incidence amongst those in whom HIV infection was unrecognized has remained unaltered at 42% [1] and mortality is highest in this group. Here, we present a case of PCP and discuss the treatment, prophylaxis, complications and current debate about the management of this disease.

Case report

A 38-year-old homosexual man was found to be HIV-antibody positive in October 1994. He was asymptomatic apart from a history of asthma. Full blood count and biochemical profile were normal and CD4 lymphocyte count was 375 cells/mm3. In early 1996, his CD4 cell count fell to 191 cells/mm3 and primary Pneumocystis prophylaxis was commenced using cotrimoxazole 960 mg day−1. He developed a transient rash of uncertain aetiology which resolved when cotrimoxazole was discontinued and did not recur upon subsequent rechallenge with the drug. He was enrolled onto a study of combination antiretroviral therapy with zidovudine (250 mg 12 hourly), lamivudine (150 mg 12 hourly) and delavirdine (a non-nucleoside reverse transcriptase inhibitor) or placebo. However, a florid rash developed 36 h after starting on combination therapy (and 2 weeks after the re-introduction of cotrimoxazole). All his medications were discontinued. The patient subsequently defaulted numerous clinic appointments and was lost to follow-up.

He presented a year later with 2 weeks of dry cough, fever (38° C) and breathlessness. Chest X-ray was typical of PCP (Figure 1) and arterial gases on air revealed hypoxia with PaO2 8.2 kPa, PaCO2 4.2 kPa, and pH 7.463. His CD4 lymphocyte count was 170 cells/mm3 and plasma viral load measured with the Roche-Amplicor® method was > 1 million copies ml−1. Full blood count, urea and electrolytes were normal. Lactate dehydrogenase activity was raised at 925 IU l−1 (normal < 620) with normal transaminases and gamma-glutamyl transferase.

Figure 1.

Figure 1

Chest X-ray of patient.

A clinical diagnosis of PCP was made and because of the possibility of allergy to cotrimoxazole, treatment was commenced with intravenous clindamycin (600 mg 6 hourly) and oral primaquine (30 mg day−1), together with oral prednisolone (50 mg day−1). Twelve days into treatment, he developed a widespread maculopapular rash and clindamycin/primaquine therapy was switched to dapsone 100 mg day−1 orally which was well tolerated. He completed 3 weeks of PCP treatment and the dose of dapsone was reduced to 50 mg day−1 as maintenance of secondary prophylaxis against PCP. At this time, antiretroviral therapy was commenced with stavudine (40 mg 12 hourly), didanosine (400 mg day−1) and indinavir (1200 mg 12 hourly).

He was readmitted 1 month later with fever (38° C) and breathlessness. On examination, he was unwell, pale, breathless and tachycardic (rate 110 min−1). There was splenomegaly. Blood pressure was 120/65 mmHg. Mycoplasma serology and blood, stool and urine cultures were negative. However, his full blood count showed anaemia with a haemoglobin of 8.8 g dl−1, white cell count 3.5×109 l−1 (neutrophils 2.77, lymphocytes 0.29), platelets 199×109 l−1 with a normal MCV. Blood film showed polychromasia, spherocytes and burr cells consistent with intravascular haemolysis. Urea was 19.9 mmol l−1, creatinine 408 μmol l−1, bicarbonate 17 mmol l−1 and lactate dehydrogenase activity was 2746 IU l−1 with normal sodium, potassium and corrected calcium levels and transaminase and creatinine kinase activities. Direct Coombs’ test and autoantibodies were negative, and methaemoglobin levels were not elevated. Ultrasound of his kidneys was normal.

Dapsone-induced haemolysis with secondary renal impairment was suspected and all his medications were stopped. Over the next 5 days his renal function and haemoglobin normalized without the need for blood transfusion. He was subsequently discharged well on his antiretroviral therapy and atovaquone syrup (500 mg 12 hourly) as PCP prophylaxis.

Discussion

Pneumocystis carinii infection only develops with significant immunosuppression (CD4 count usually < 200/cell mm3) [2]. It is uncertain if the majority of infections are newly acquired or subsequent reactivation of previous infections [3].

Paradoxically, starting antiretroviral therapy in previously untreated patients with a low CD4 count may unmask PCP, presumably due to restoration of immunological responses to subclinical Pneumocystis colonization [4]. In these patients, antiretroviral therapy should be discontinued until the PCP has been adequately treated [5]. This is not recommended for patients who develop PCP having been on a successful highly active antiretroviral therapy regimen for some time.

Typical clinical features are breathlessness (usually of insidious onset), fever and nonproductive cough (the presence of sputum makes PCP less likely). Patients with previously undiagnosed HIV infection will usually also have oral candidiasis [6]. Minimal signs are present on chest auscultation, except in severe disease where coarse, dry crackles are heard. Chest X-ray characteristically shows diffuse alveolar (ground glass) shadowing such as seen our case (Figure 1), often starting in peri-hilar areas. Effusions are rare, but spontaneous pneumothorax may occur. Serum lactate dehydrogenase is elevated and may be used in conjunction with blood gases and chest X-ray to monitor the progress of therapy.

Differential diagnoses include bacterial pneumonia (Haemophilus spp., Streptococcus pneumoniae, ‘atypical’ infections such as Mycoplasma, Legionella and Chlamydia, mycobacterial infections, Nocardia), fungal infections (Cryptococcus, Histoplasma, Coccidioides) and noninfectious causes (pulmonary oedema, pulmonary Kaposi’s sarcoma, lymphoma).

Diagnosis

The combination of HIV-seropositivity, a CD4 count <200/mm3, typical radiological appearance and hypoxia, often only on exertion, is sufficient for a diagnosis and empirical treatment can be started. If in doubt, bronchoalveolar lavage should be performed [3]. Induced sputum is less sensitive and a negative result does not exclude the diagnosis. Confirmation of the diagnosis is made histologically (immunofluorescence, a variety of silver stains or Giemsa [6]). Histopathological diagnosis is still possible up to 4 days after commencing the treatment. DNA detection methods have a much higher sensitivity but still remain a research tool.

Treatment

Severity is based upon the degree of hypoxia [7]. This may be assessed by PaO2 on air, alveolar-arterial gradient, degree of desaturation on exercise or alternatively using the hypoxaemic ratio in patients receiving supplementary oxygen (PaO2 [in mmHg], divided by fraction of inspired oxygen) [8]. Severe PCP is defined as: severe hypoxia (arterial PaO2<8.0 kPa on air), or a hypoxaemic ratio of (< 250). These patients usually have extensive chest X-ray changes and dyspnoea at rest. Moderate PCP is defined as: PaO2 between 8.1 and 11.0 kPa on air, or a hypoxaemic ratio of 250–350. In moderate to severe disease, there is (>5% oxygen desaturation on exercise or an alveolar-arterial gradient >4.7 kPa on room air [3]. Mild PCP is defined as PaO2 on air>11.0 kPa or a hypoxaemic ratio >350.

Corticosteroids:

Corticosteroids should be used for moderate to severe PCP and commenced together with anti-Pneumocystis therapy. They increase survival and prevent the development of acute respiratory failure [911]. Some authors suggest that their use in moderate cases should be for PaO2 (9.3 kPa or an alveolar–arterial pressure of 4.7 kPa) [3]. Their use in ventilated patients has also reduced mortality from 84% to 39% [11]. The recommended doses are oral prednisolone 50–80 mg day−1 for at least the first 5 days, tapering off over the next 2–3 weeks [12]. In severe disease, intravenous hydrocortisone should initially be used.

Anti-Pneumocystis therapy:

(Table 1). All treatment courses are for 21 days (shorter courses of 14 days may be associated with treatment failure). The drug of first choice is cotrimoxazole (trimethoprim-sulphamethoxazole) at a dose of 120 mg kg−1day−1 in four divided doses. Patients with moderate to severe disease should initially receive the drug intravenously, as absorption may be impaired [6] and clinical deterioration may occur in the first 5 days of treatment.

Table 1.

Summary of first and second line PCP treatments. (All treatment courses for 3 weeks).

graphic file with name bcp0047-0605-t1.jpg

Side-effects are common in HIV infected patients. The most frequent are rash which at the high doses used in treatment is seen in ≈50% of cases and is often severe and associated with fever, hepatitis (20%), leukopenia (15%) and thrombocytopenia (15%). Deteriorating renal function should prompt the examination of the urine for eosinophils, a marker of interstitial nephritis. Concomitant steroids may reduce or delay the development of cutaneous reactions [13]. Mild skin rashes are not an indication for stopping the treatment and tend to resolve spontaneously [6].

Alternatives to cotrimoxazole: Patients intolerant of cotrimoxazole:

Intravenous pentamidine is as effective as cotrimoxazole but with higher toxicity [14, 15], and is given at a dose of 4 mg kg−1. Pentamidine achieves therapeutic levels in the lung slowly (5–7 days) due to high levels of extrapulmonary tissue binding [9]. Severe hypotension may develop with the first dose whilst hypo-or hyperglycaemia may appear at any time during or after therapy. Other side-effects include cardiac dysrhythmias, renal failure, pancreatitis, myoglobinaemia and elevation of creatinine kinase levels [16, 17]. The dose interval should be increased in the event of renal impairment (creatinine clearance < 35 ml min−1).

Clindamycin

(450–600 mg 6 hourly, oral or i.v.) plus primaquine(30 mg day−1 orally) has comparable efficacy with cotrimoxazole in mild to moderate disease [17]. Side-effects include skin rashes and antibiotic-associated diarrhoea. Primaquine can cause methaemoglobinaemia and haemolysis in patients with G6PD deficiency.

Dapsone

(100 mg day−1) plus trimethoprin (20 mg kg−1day−1) is a suitable alternative for mild to moderate cases. Side-effects include rash and haemolytic anaemia. Although the side-effects of cotrimoxazole are usually ascribed to sulphamethoxazole, many reactions may also be due to the trimethoprim moiety.

Oral atovaquone

is better tolerated than cotrimoxazole [18] but in clinical trials has shown higher failure rates in part due to its unreliable absorption. A new liquid suspension with improved bioavailability has been introduced, but whether efficacy is improved remains unanswered. The new dosing regimen is 750 mg (5 ml) twice daily for mild to moderate disease. The drug should be taken with meals (preferably with a high fat content) to enhance absorption.

Patients failing cotrimoxazole:

If deterioration occurs despite intravenous cotrimoxazole, complications such as pulmonary oedema, superinfection, pneumothorax or other opportunistic infections should be excluded, as in the differential diagnosis above. Patients failing cotrimoxazole have a poor prognosis. Controlled trials are largely lacking, but alternatives are: (i) addition or substitution with intravenous pentamidine (ii) addition of oral dapsone or atovaquone (iii) substituting with oral clindamycin and primaquine, or (iv) salvage therapy with trimetrexate and folinic acid (leucovorin) rescue (see Table 1).

Methods of respiratory support:

The goal of therapy is correcting the hypoxaemia. High flow oxygen should be given via standard masks and oxygen saturation should be monitored closely in moderate to severe disease.

When adequate oxygen saturation is not achieved on maximum FiO2 continuous positive airway pressure (CPAP) should be considered [19]. If this fails mechanical ventilation should be offered to the patient. This carries a high mortality [20] but in view of the long-term outcome of HIV infection in the era of highly active antiretroviral therapy (HAART) should not be denied.

Prophylaxis

Prophylaxis may be instituted once CD4 counts drop below 200/mm3 (primary prophylaxis) or following PCP (secondary prophylaxis). The first line agent is oral cotrimoxazole at a dose of 960 mg day−1 (daily, or 3 times a week) which also provides prophylaxis against toxoplasmosis. Alternatives is nebulized pentamidine (300–600 mg, 2–4 weekly) but this is not as effective [21], confers no protection against extra-pulmonary Pneumocystis infection and is less tolerated in patients with bronchospasm; dapsone (50–100 mg day−1); or atovaquone suspension which has been recently shown to be as effective as dapsone and better tolerated [22] (1500 mg daily).

Discontinuation of prophylaxis

The success of combination antiretroviral treatment in elevating CD4 counts in many patients has led to debate regarding whether to discontinue PCP prophylaxis when CD4 count are restored. Several studies are now in progress and preliminary data suggest that withdrawal of prophylaxis once counts of > 200/mm3 are achieved is safe (although duration of follow-up is as yet short) [23, 24], and this is our practice. Further studies are required to address whether this is also safe for secondary prophylaxis, and whether the CD4 nadir should be taken into consideration.

Conclusion

PCP still carries a high mortality, especially when respiratory failure develops. Mortality is highest amongst patients in whom HIV is unsuspected and who consequently present late. Prompt diagnosis and treatment, including early use of steroids have been shown to reduce mortality. Combination antiretroviral therapy has altered the natural history of AIDS and patients who survive PCP are likely to have considerably improved life expectancy.

Learning points

  • If Pneumocystis carinii pneumonia is suspected clinically, treatment should be started promptly.

  • Prophylaxis against Pneumocystis carinii pneumonia should be initiated when the CD4 count is <200 cells per mm3. Anti-viral chemotherapy (HAART) has significantly improved the survival of patients with AIDS even with late stage disease. Pneumocystis carinii pneumonia is not an end stage event.

  • Corticosteroids are life saving in severe Pneumocystis carinii pneumonia.

  • Patients can deteriorate rapidly during treatment. Close monitoring of respiratory rate and arterial oxygenation are essential.

  • Breathlessness is the predominant sympton of Pneumocystis carinii pneumonia; there are often no signs on examination of the chest.

References

  • 1.PHLS. Communicable Disease Surveillance Centre. AIDS and HIV infection in the United Kingdom: monthly report. Communicable Dis Report. 1998;8:155–158. [PubMed] [Google Scholar]
  • 2.Masur H, Ognibene FP, Yarchoan R, et al. CD4 counts as predictors of opportunistic pneumonias in human immunodeficiency virus (HIV) infection. Ann Intern Med. 1989;111:223. doi: 10.7326/0003-4819-111-3-223. [DOI] [PubMed] [Google Scholar]
  • 3.Miller RF, Mitchell DM. AIDS and the lungs. Update 1995. Thorax. 1995;50:191–200. doi: 10.1136/thx.50.2.191. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Kandil O, Fishman JA, Koziel H, Pinkston P, Rose RM, Remold HG. Human immunodeficiency virus type 1 infection of human macrophages modulates the cytokine response to Pneumocystis carinii. Infect Immun. 1994;62:644–650. doi: 10.1128/iai.62.2.644-650.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Rodriguez-Rosado R, Soriano V, Dona C, Jimenez-Nacher I, Gonzalez-Lahoz J. Opportunistic infections shortly after beginning highly active antiretroviral therapy (HAART) AIDS. 1998;12:S87. [PubMed] [Google Scholar]
  • 6.Vilar FJ, Wilkins EGL. Management of Pneumocystis carinii pneumonia in intensive care. Br J Intensive Care. 1998;8:53–62. [Google Scholar]
  • 7.Miller R. Prevention and treatment of Pneumocystis carinii pneumonia in patients infected with HIV. Drug Therap Bull. 1994;32:12–15. doi: 10.1136/dtb.1994.32212. [DOI] [PubMed] [Google Scholar]
  • 8.Bozzette SA, Sattler FR, Chiu JC, et al. A controlled trial of early adjunctive treatment with corticosteroids for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. N Engl J Med. 1990;323:1451–1457. doi: 10.1056/NEJM199011223232104. [DOI] [PubMed] [Google Scholar]
  • 9.Fishman JA. Treatment of infection due to Pneumocystis carinii. Antimicrobial Agents Chemotherapy. 1998;42:1309–1314. doi: 10.1128/aac.42.6.1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Montaner JSG, Lawson LM, Levitt N, Belzberg A, Schechter MT, Ruedy J. Corticosteroids prevent early deterioration in patients with moderately severe Pneumocystis carinii pneumonia and the acquired immunodeficiency syndrome. Ann Intern Med. 1990;113:14–20. doi: 10.7326/0003-4819-113-1-14. [DOI] [PubMed] [Google Scholar]
  • 11.Montaner JSG, Russel JA, Lawson LM, Ruedy J. Acute respiratory failure secondary to Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. Chest. 1989;95:881–884. doi: 10.1378/chest.95.4.881. [DOI] [PubMed] [Google Scholar]
  • 12.Sistek CL, Wordell CJ, Hauptman SP. Adjuvant corticosteroid therapy for Pneumocystis carinii pneumonia in AIDS patients. Ann Pharmacother. 1992;26:1127–1133. doi: 10.1177/106002809202600915. [DOI] [PubMed] [Google Scholar]
  • 13.Caumes E, Roudier C, Rogeaux O, Bricaire F, Gentilini M. Effect of corticosteroids on the incidence of adverse cutaneous reaction to trimethoprim-sulphamethoxazole during treatment of AIDS-associated Pneumocystis carinii pneumonia. Clin Infect Dis. 1994;18:319–323. doi: 10.1093/clinids/18.3.319. [DOI] [PubMed] [Google Scholar]
  • 14.Sattler FR, Cowan R, Nielsen DM, Ruskin J. Trimethoprim-sulphamethoxazole compared with pentamidine for treatment of Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. Ann Intern Med. 1988;109:280–287. doi: 10.7326/0003-4819-109-4-280. [DOI] [PubMed] [Google Scholar]
  • 15.Brooks KR, Ong R, Spector RS, Greenbaum DM. Acute respiratory failure due to Pneumocystis carinii pneumonia. Crit Care Clinics. 1993;9:31–48. [PubMed] [Google Scholar]
  • 16.Hopewell PC, Masur H. Pneumocystis carinii pneumonia: current concepts. In: Sande MA, Volberding MD, editors. The Medical Management of AIDS. Philadelphia: W.B. Saunders, Company; 1995. pp. 367–401. [Google Scholar]
  • 17.Toma E, Fournien S, Dumond M, Boldue P, Deschamps H. Clindamycin/primaquine versus trimethoprim-sulphamethoxazole as primary therapy for Pneumocystis carinii pneumonia in AIDS. a randomised, double blind pilot trial. Clin Infect Dis. 1993;17:178–184. doi: 10.1093/clinids/17.2.178. [DOI] [PubMed] [Google Scholar]
  • 18.Hughes W, Leoung G, Kramer F, et al. Comparison of atovaquone (566C80) with trimethoprim-sulphamethoxazole to treat Pneumocystis carinii pneumonia in patients with AIDS. N Engl J Med. 1993;328:1521–1527. doi: 10.1056/NEJM199305273282103. [DOI] [PubMed] [Google Scholar]
  • 19.Staikowsky F, Lafon B, Guidet B, Denis M, Mayaud C, Offenstadt G. Mechanical ventilation for pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome. Is the prognosis really improved? Chest. 1993;104:756–762. doi: 10.1378/chest.104.3.756. [DOI] [PubMed] [Google Scholar]
  • 20.Dohn MN, Baughmann RP, Vigdorth EM, Frame DL. Equal survival rates for first, second, and third episodes of Pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome. Arch Intern Med. 1992;152:2465–2470. doi: 10.1001/archinte.152.12.2465. [DOI] [PubMed] [Google Scholar]
  • 21.Hardy WD, Feinberg J, Filkelstein DM, et al. A controlled of trimethoprim-sulphamethoxazole or aerosolised pentamidine for secondary prophylaxis of Pneumocystis carinii pneumonia in patients with the acquired immunodeficiency syndrome. N Engl J Med. 1992;327:1842–1848. doi: 10.1056/NEJM199212243272604. [DOI] [PubMed] [Google Scholar]
  • 22.El-Sadr WM, Murphy RL, McCabe Yurik T, et al. Atovaquone compared with dapsone for the prevention of Pneumocystis carinii pneumonia in patients with HIV infection who cannot tolerate trimethoprim, sulphonamides, or both. N Engl J Med. 1998;339:1889–1895. doi: 10.1056/NEJM199812243392604. [DOI] [PubMed] [Google Scholar]
  • 23.Reiss P, Weverling GJ, Mocroft A, et al. Conference Record 12th World AIDS Conference, June 28–July 3. Geneva, Switzerland: 1998. Can prophylaxis for PCP and cerebral toxoplasmosis be safely discontinued in HIV-infected patients successfully treated with HAART? The EuroSIDA Study. [Google Scholar]
  • 24.Furrer H, Opavil M, Bernasconi E, et al. Conference Record. 12th World AIDS Conference, June 28–July 3. Geneva, Switzerland: 1998. Is it safe to stop primary PCP prophylaxis in patients treated with anti-retroviral combination regimens? [Google Scholar]

Articles from British Journal of Clinical Pharmacology are provided here courtesy of British Pharmacological Society

RESOURCES