The introduction of potent antiretroviral therapy in HIV-infection has lead to an increase in CD4-positive T lymphocytes, to the suppression of plasma virus concentration, often to undetectable levels, and to a substantial decline in HIV-associated morbidity and mortality [1–3]. The question as to whether or not treatment-associated restoration of the immune response allows prophylaxis against Pneumocystis carinii pneumonia (PCP) to be discontinued has been examined in several cohort studies. Findings from the Swiss HIV Cohort Study , the EuroSIDA Study  and other prospective studies  showed that discontinuation of primary prophylaxis against PCP may be safe in patients who respond to potent antiretroviral therapy with a sustained increase in CD4 cell count. The preliminary results from an ongoing randomized trial confirm these findings [7,8]. However, follow-up in these studies was short (typically 1 year or shorter) and concern about the long-term safety of stopping prophylaxis therefore remains. There is also concern about discontinuation in patients who experienced very low CD4 cell counts before initiating antiretroviral combination therapy, and in patients in whom suppression of plasma virus concentration was incomplete when stopping prophylaxis [9–11]. For example, the 1999 USPHS/IDSA guidelines  state that ‘additional criteria (for discontinuation) might include sustained reduction in viral load for at least 3–6 months.'
The present study extends a previous analysis of the Swiss HIV Cohort Study  to a larger patient group with longer follow-up and focuses on groups where concern remains about the safety of discontinuing prophylaxis [9–11] : patients who experienced very low CD4 cell counts before initiating antiretroviral combination therapy, and patients in whom suppression of plasma virus concentration was incomplete when stopping prophylaxis.
Patients and methods
Described in detail elsewhere, the Swiss HIV Cohort Study (SHCS) is a prospective cohort study of adult HIV-infected patients [1,13]. Patients are followed in one of seven study centres (Basle, Berne, Geneva, Lausanne, Lugano, St Gall, and Zurich). Enrolment is independent of stage of disease or degree of immunosuppression, and information is collected according to standardized criteria on structured forms at enrolment and at follow-up visits at 6 monthly intervals. Clinical stage is defined according to the 1993 classification system for HIV infection of the Centers for Disease Control and Prevention (CDC) . CD4 cell counts are measured using flow-cytometric assays and plasma HIV-RNA levels are measured using the Roche Amplicor assay with a lower detection limit of about 200 copies/ml. An ultrasensitive modification of the assay with a limit of detection of 2–40 copies/ml was introduced during 1998 .
SHCS participants on primary prophylaxis against PCP were eligible if they responded to antiretroviral combination therapy with an increase in their CD4 cell counts to at least 200 × 106/l and 14% of peripheral lymphocyte count and if the CD4 cell count remained above these levels for at least 12 weeks. The ethics committees of all seven centres approved the study and written informed consent was obtained from all patients. According to the protocol primary PCP prophylaxis would be resumed if the CD4 cell count fell below one of the threshold values on two consecutive measurements. Recruitment started in June 1997, and ended in October 1999.
Definitions of high risk groups and endpoints
Two groups of patients at potentially high risk of developing PCP were defined. The first group included all patients whose lowest ever measured CD4 cell count was below 100 × 106/l. The second group included patients whose plasma HIV-1 RNA at study entry was above 200 copies/ml. The two groups were not mutually exclusive.
A diagnosis of PCP was the primary endpoint of this study: A diagnosis of PCP was made: if P. carinii was found microscopically in induced sputum or bronchoalveolar fluid or in a histological lung specimen; or if there was a history of dyspnoea on exertion or non-productive cough in the absence of evidence of bacterial pneumonia and the patient responded to standard PCP treatment. CDC clinical stage B and C events and treatments with antibiotics were secondary endpoints.
It was decided a priori that discontinuation of primary prophylaxis would be considered to be safe if the upper 95% confidence limit of the incidence was below the upper 95% confidence limit recorded in a historical cohort of 1277 patients who were followed up in the SHCS with CD4 cell counts of 200–500 × 106/l in the period 1990–1994, before potent therapies became available. In this patient group primary prophylaxis against PCP is not indicated according to current guidelines . The incidence of PCP in this group was 1.1 per 100 person-years [95% confidence interval (CI), 0.6–1.8].
The closing date for the present analysis was December 31 1999. In January 2000 all charts were reviewed and patients were contacted by telephone if the treating physician had not seen them within the previous month. The length of follow-up was measured from the date that primary prophylaxis was discontinued to the date of last contact, the date of reaching an endpoint or the date of withdrawal. Events were assumed to have a Poisson distribution, and exact 95% CI were calculated for the incidence of endpoints. One-sided upper 95% confidence limits were calculated if no event was recorded. Stata software (version 6.0; College Station, Texas, USA) was used for statistical analysis.
A total of 491 patients followed up within the SHCS fulfilled the entry criteria. As of October 1999, 396 (80.1%) of them had been enrolled. There were no statistically significant differences (P > 0.05) between the patients enrolled and not enrolled with respect to age, sex, transmission mode, clinical stage, or nadir CD4 cell count.
In 336 patients (84.8%) the discontinued prophylactic regimen consisted of cotrimoxazole. The median duration of prophylaxis was 26 months. Characteristics at baseline and at study end are shown in Table 1. The subgroup of patients whose lowest ever measured CD4 cell count was below 100 × 106/l consisted of 191 participants, and the subgroup of patients whose plasma HIV-1 RNA at study entry was above 200 copies/ml consisted of 144 patients. Sixty-one participants were included in both subgroup analyses and 122 participants were not included in any subgroup analysis.
The distribution across sex, age and transmission categories was similar in the different patient groups. Between one-quarter and one-third of patients had a history of an AIDS-defining illness. About half of the patients (187, 47.2%) had a history of a condition known to be associated with PCP (oral thrush, Candida esophagitis, unexplained weight loss, unexplained fever, wasting syndrome or recurrent bacterial pneumonia) [16,17]. In 30 patients (7.6%) follow-up was censored before the closing date: 16 resumed prophylaxis because their CD4 cell count fell below threshold values, two resumed prophylaxis for recurrent lower and upper respiratory tract infections, six patients could not be located and two patients withdrew consent. Four patients died: one of hepatocellular carcinoma, one of non-Hodgkin's Lymphoma, one of heroin overdose and one of sudden death, probably cardiac arrest.
Incidence of PCP
One patient was diagnosed with PCP. He presented with chronic cough for several weeks without dyspnoea, fever or decrease in general performance. On clinical examination he had prolonged expiration and wheezing. His chest X-ray was considered normal. P. carinii was detected by immunofluorescence in a smear of spontaneous sputum. He was treated with cotrimoxazole for 3 weeks and the cough resolved. The CD4 cell count at diagnosis was 530 × 106/l and plasma HIV RNA was 38 copies/ml. He had started primary prophylaxis with cotrimoxazole 18 months earlier with a nadir cell CD4 count of 250 × 106/l because of Candida oesophagitis and recurrent oral thrush and discontinued prophylaxis 5 months before the diagnosis was made when his cell CD4 count was 302 × 106/l and plasma HIV RNA was undetectable.
The overall incidence of PCP was thus 0.18 (95% CI, 0–1.0 per 100 person-years). The patient with a PCP diagnosis was in neither of the two subgroups at potentially increased risk, so that the incidence in these subgroups was 0 per 100 person-years (Table 2). In all instances the upper 95% confidence limits were below the predefined threshold incidence of 1.8 per 100 patient years.
Other AIDS events and bacterial infections
Five patients experienced a CDC C event other than PCP (two Candida oesophagitis, two non-Hodgkin's lymphoma, one cryptosporidiosis) and 10 patients presented with a CDC B event (three multidermatomal herpes zoster, two cervical dysplasia, two thrombocytopenia, two oral hairy leukoplakia and one unexplained weight loss). Sixty-eight patients (17%) had a total of 81 courses of systemic antibiotic treatment for the following conditions: 38 respiratory infections such as sinusitis, pharyngitis, otitis media and bronchitis, 17 typical or atypical pneumonia, 10 skin or soft tissue infections, eight urinary tract infections, five diarrhoea, two sexually transmitted diseases, and one sepsis with Staphylococcus aureus. The incidence of pneumonia was 3.1 episodes per 100 person-years (95% CI, 1.8–4.9 per 100 person-years). The incidence of pneumonia in intravenous drug users was similar: 4.2 per 100 person-years (95% CI, 1.6–8.6 per 100 person-years).
The discontinuation of primary prophylaxis against PCP in HIV-infected patients responding to potent antiretroviral therapy could improve the quality of life of patients, reduce the risks associated with adverse effects and drug interactions, and decrease costs to health services. But when can primary prophylaxis be discontinued safely? The present study extends a previous analysis of the SHCS  to a larger patient group with longer follow-up and gives special consideration to the experience of patients perceived to be at increased risk. To our knowledge this study represents the largest database of patients followed prospectively after discontinuing primary prophylaxis against PCP. Our results indicate that it is safe to discontinue prophylaxis in patients who responded with a sustained increase in CD4 cell count to at least 200 × 106/l and 14% of the total lymphocyte count, irrespective of the CD4 cell count nadir and the viral load at the time of stopping prophylaxis. We showed recently that this is true also for patients on primary prophylaxis who are seropositive for IgG against Toxoplasmosis gondii .
High viral load has been shown to be an independent predictor of progression to AIDS , including PCP , in the era before potent antiretroviral drugs became available. The restoration of some immune functions in vivo depends on the rate of viral load decline during the initial phase of antiretroviral combination therapy . However, our findings suggest that a detectable viral load does not impair Pneumocystis-specific immune response if the CD4 cell count has increased above the threshold levels described. This finding is in agreement with the sustained immunological and clinical benefit observed in patients treated with antiretroviral combination therapy whose viraemia is suppressed suboptimally [22–25].
In patients who achieved CD4 cell count level of more than 200 × 106/l while on antiretroviral combination therapy the incidence of new opportunistic infections was dependent on the nadir CD4 cell count before therapy was started, as shown in a recently published analysis of the EuroSIDA cohort . The hazard of new AIDS-defining infections was more than three times higher in patients with CD4 cell count nadirs below 100 × 106/l as compared to those with nadirs above 150 × 106/l. However the median follow-up was only 4 months in this study, indicating that most opportunistic infections occurred during the first months after reaching the CD4 threshold mentioned above. Our results indicate that the CD4 cell count nadir that was reached before starting potent combination therapy is irrelevant for the decision to discontinue prophylaxis once a CD4 cell count of at least 200 × 106/l and 14% of total lymphocytes is maintained for at least 12 weeks.
PCP was diagnosed in one patient who was treated successfully with antiretroviral combination therapy. His CD4 cell count was above 500 × 106/l and his plasma HIV RNA was below 50 copies/ml at the time the diagnosis was made. Although the clinical picture was atypical, P. carinii was found in the sputum and the chronic cough subsided after treatment. This raises a question about the spectrum of clinical signs and symptoms, and whether subclinical infections or reactivations occur in patients treated successfully with antiretroviral combination therapy. P. carinii may be found in body fluids of immunocompromised patients who do not progress to clinically overt PCP . This possibility should be taken into account in future studies of PCP in HIV-infected patients treated with potent antiretroviral therapy.
Our study has a number of limitations. Firstly, it was an uncontrolled, observational study. Selection bias could have been introduced, for example if patients at low risk were included preferentially. Such bias is unlikely for several reasons. The SHCS is a study with national coverage and includes approximately 70% of patients with advanced HIV disease in the country . A large proportion of the eligible patients was included, and those who were not included had characteristics similar to those who were not. The results from a randomized trial from Spain, presented at recent conferences, have shown zero events in the discontinuation arm . Second, although the median follow-up was 18 months and 75 patients were followed-up for longer than 2 years our study cannot exclude with certainty an increased risk of P. carinii pneumonia several years after stopping prophylaxis. This seems unlikely, however, as immune function increases with time after initiation of combination therapy . An earlier analysis of the Swiss cohort showed that the incidence of opportunistic infections declines rapidly 3 months after starting therapy . Third, our results cannot be extended to patients who do not respond to potent therapy with a sustained increase of the CD4 cell count. Finally, we did not study secondary prophylaxis, but a recent retrospective analysis of eight European cohorts indicated that discontinuation of secondary PCP prophylaxis may also be safe .
Our findings are in agreement with other observational studies examining the risk of discontinuing primary prophylaxis against P. carinii, as summarized in Table 3. There are two studies with a prospective design similar to this, one from the Netherlands  and one from Denmark . Combining the data from the present study with the two previous studies gives a total of 651 patients, 777 person-years of follow-up and two cases of PCP, an incidence of 0.26 per 100 person-years (95% CI, 0.03–0.94 per 100 person-years). In addition there are two analyses of cohort studies, one from Europe  and one from the USA . No case of PCP was identified in these studies (Table 3). As more and more data become available about discontinuation of primary and secondary prophylaxis against opportunistic infections a collaboration by the different research groups to combine the results should be attempted.
Chronic treatment with cotrimoxazole may prevent other bacterial infections in HIV-infected patients, especially bacterial pneumonia. The incidence of pneumonia observed in our study (3.1 episodes per 100 person-years) was lower than that observed in an American study  before antiretroviral combination therapy became available. In the latter study the incidence of pneumonia was 6.8 per 100 person-years among patients with CD4 cell counts between 200 and 500 × 106/l. Given the increased risk of development of bacterial resistance against cotrimoxazole  during prophylaxis and the low fatality rate of bacterial pneumonia, prophylaxis of bacterial pneumonia cannot be justified in this population.
We conclude that it is safe to stop primary PCP prophylaxis in patients who have responded to combination antiretroviral therapy with a sustained increase in the CD4 cell count to at least 200 × 106/l and 14% of their total peripheral lymphocyte count, irrespective of CD4 cell count nadir and viral load at the time of stopping prophylaxis.
We are indebted to our patients who gave their consent for enrolment into this study knowing that there might be a risk of acquiring opportunistic infections. We thank the physicians and study nurses of the SHCS centres and the general practitioners who cared for the patients in the study. We thank especially A. Christen, study nurse in Berne, who co-ordinated data collection, and N. Low and P. Vernazza who made useful comments on earlier drafts of the manuscript.
1. Egger M, Hirschel B, Francioli P. et al
. Impact of new antiretroviral combination therapies in HIV infected patients in Switzerland: prospective multicentre study.
BMJ 1997, 315: 1194 –1199.
2. Palella FJ, Delaney KM, Moorman AC. et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection.
N Engl J Med 1998, 338: 853 –860.
3. Detels R, Munoz A, McFarlane G. et al
. Effectiveness of potent antiretroviral therapy on time to AIDS and death in men with known HIV infection duration.
:Multicenter AIDS Cohort Study Investigators [see comments].
JAMA 1998, 280: 1497 –1503.
4. Furrer H, Egger M, Opravil M. et al
. Discontinuation of primary prophylaxis against Pneumocystis carinii pneumonia in HIV-1 infected adults treated with combination antiretroviral therapy.
N Engl J Med 1999, 340: 1301 –1306.
5. Weverling GJ, Mocroft A, Ledergerber B. et al
. Discontinuation of Pneumocystis carinii pneumonia prophylaxis after start of highly active antiretroviral therapy in HIV-1 infection.
:EuroSIDA Study Group.
Lancet 1999, 353: 1293 –1298.
6. Schneider MME, Borleffs JCC, Stolk RP, Jaspers AJJ, Hoepelman AIM. Discontinuation of prophylaxis for Pneumocystis carinii pneumonia in HIV-1-infected patients treated with highly active antiretroviral therapy.
Lancet 1999, 353: 201 –203.
7. Lopez JC, Pena JM, Miro JM, Podzamczer D. Discontinuation of PCP prophylaxis (PRO) is safe in HIV-infected patients (PTS) with immunological recovery with HAART. Preliminary results of an open, randomised and multicentric clinical trial (GESIDA04/98). Sixth Conference on Retroviruses and Opportunistic Infections
. Chicago, January–February 1999 [abstract LB7].
8. Miro JM, Lopez JC, Podzamczer D, et al. Discontinuation of toxoplasmic encephalitis prophylaxis is safe in HIV-1 and T. gondii coinfected patients after immunological recovery with HAART: preliminary results from the GESIDA 04/98-B Study. Seventh Conference on Retroviruses and Opportunistic Infections.
San Francisco, January–February 2000 [abstract 230].
9. Masur H, Kaplan J. Does Pneumocystis carinii prophylaxis still need to be lifelong? [editorial comment].
N Engl J Med 1999, 340: 1356 –1358.
10. Feinberg J. Withdrawal of prophylaxis against Pneumocystis carinii pneumonia [comment].
Lancet 1999, 353: 1287. 1287.
11. Furrer H, Jacobson MA, Reiss P, Telenti A. Discontinuation of prophylaxis against opportunistic infections in HIV-infected persons receiving potent combination antiretroviral therapy.
AIDS Rev 1999, 1: 183 –192.
12. Centers for disease control and prevention. 1999 USPHS/IDSA guidelines for the prevention of opportunistic infections in persons infected with human immunodeficiency virus.
:U.S. Public Health Service (USPHS) and Infectious Diseases Society of America (IDSA).
MMWR 1999, 48 (RR-10): 1 –66.
13. Ledergerber B, von Overbeck J, Egger M, Lüthy R. The Swiss HIV cohort study: rationale, organization and selected baseline characteristics.
Soz Präventivmed 1994, 39: 387 –394.
14. Centers for disease control. 1993 Classification system for HIV infection and expanded surveillance case definition for acquired immunodeficiency syndrome (AIDS) among adolescents and adults.
MMWR 1992, 41 (RR-17): 1 –19.
15. Schockmel GA, Yerly S, Perrin L. Detection of low HIV-1 RNA levels in plasma.
J Acquir Immune Defic Syndr Hum Retrovirol 1997, 14: 179 –183.
16. Phair JP, Munoz A, Detels R. et al
. The Risk of Pneumocystis carinii pneumonia among men with human immunodeficiency virus type 1.
N Engl J Med 1990, 322: 161 –165.
17. Kaplan JE, Hanson DL, Navin TR, Jones JL. Risk factors for primary Pneumocystis carinii pneumonia in human immunodeficiency virus-infected adolescents and adults in the United States: reassessment of indications for chemoprophylaxis.
J Infect Dis 1998, 178: 1126 –1132.
18. Furrer H, Opravil M, Bernasconi E, Telenti A, Egger M, for the Swiss HIV Cohort Study
. Stopping primary prophylaxis in HIV-1 infected patients at high risk of Toxoplasma encephalitis.
Lancet 2000, 335: 2217 –2218.
19. Mellors JW, Munoz A, Giorgi JV. et al
. Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection.
Ann Intern Med 1997, 126: 946 –954.
20. Williams PL, Currier JS, Swindells S. Joint effects of HIV-1 RNA levels and CD4 lymphocyte cells on the risk of specific opportunistic infections.
AIDS 1999, 13: 1035 –1044.
21. Wendland T, Furrer H, Vernazza P. et al
. HAART in HIV-infected patients: restoration of antigen-specific CD4 T-cell responses in vitro is correlated to CD4-memory T-cell reconstitution, whereas improvement in delayed type hypersensitivity is related to a decrease in viremia.
AIDS 1999, 13: 1857 –1862.
22. Kaufmann D, Pantaleo G, Sudre P, Telenti A, for the Swiss HIV Cohort Study
. CD4-cell count in HIV-1-infected individuals remaining viraemic with highly active antiretroviral therapy (HAART).
Lancet 1998, 351: 723 –724.
23. Ledergerber B, Egger M, Opravil M. et al
. Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: a prospective cohort study.
Lancet 1999, 353: 863 –868.
24. Piketty C, Castiel P, Belec L. et al
. Discrepant responses to triple combination antiretroviral therapy in advanced HIV disease.
AIDS 1998, 12: 745 –750.
25. Ledergerber B, Egger M, Erard V. et al
. AIDS-related opportunistic illnesses occurring after initiation of potent antiretroviral therapy: the Swiss HIV Cohort Study.
JAMA 1999, 282: 2220 –2226.
26. Miller V, Mocroft A, Reiss P. et al
. Relations among CD4 lymphocyte count nadir, antiretroviral therapy, and HIV-1 disease progression: results from the EuroSIDA study.
Ann Intern Med 1999, 130: 570 –577.
27. Leigh TR, Wakefield AE, Peters SE, Hopkin JM, Collins JV. Comparison of DNA amplification and immunofluorescence for detecting Pneumocystis carinii in patients receiving immunosuppressive therapy.
Transplantation 1992, 54: 468 –470.
28. Gebhardt M, Rickenbach M, Egger M, and the Swiss HIV Cohort Study
. Impact of antiretroviral combination therapies on AIDS surveillance reports in Switzerland.
AIDS 1998, 12: 1195 –1201.
29. Pakker NG, Notermans DW, De Boer RJ. et al
. Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection: A composite of redistribution and proliferation.
Nature Med 1998, 4: 209 –214.
30. Ledergerber B, Mocroft A, Reiss P, et al. It is safe to discontinue secondary prophylaxis for PCP in HIV-infected patients treated with HAART: results from eight prospective European cohorts. Seventh Conference on Retroviruses and Opportunistic Infections.
San Francisco, January–February 2000 [abstract LB5].
31. Kirk O, Lundgren JD, Pedersen C, Nielsen H, Gerstoft J. Can chemoprophylaxis against opportunistic infections be discontinued after an increase in CD4 cells induced by highly active antiretroviral therapy?
AIDS 1999, 13: 1647 –1651.
32. Yangco BG, Von Bargen JC, Moorman AC, Holmberg SD. Discontinuation of chemoprophylaxis against Pneumocystis carinii pneumonia in patients with HIV infection.
:HIV Outpatient Study (HOPS) Investigators.
Ann Intern Med 2000, 132: 201 –205.
33. Hirschtick RE, Glassroth J, Jordan MC. et al
. Bacterial pneumonia in persons infected with the human immunodeficiency virus.
N Engl J Med 1995, 333: 845 –851.
34. Martin JN, Rose DA, Hadley WK, Perdreau-Remington F, Lam PK, Gerberding JL. Emergence of trimethoprim-sulfamethoxazole resistance in the AIDS era.
J Infect Dis 1999, 180: 1809 –1818.
Members of the Swiss HIV Cohort Study: M. Battegay (Chairman of the Scientific Board), E. Bernasconi, Ph. Bürgisser, M. Egger, P. Erb, W. Fierz, M. Flepp (Chairman of the Clinical and Laboratory Committee), P. Francioli (President of the SHCS, Centre Hospitalier Universitaire Vaudois, Lausanne), H. Furrer, P. Grob, B. Hirschel, B. Ledergerber, L. Matter, A. Meynard, M. Opravil, F. Paccaud, G. Pantaleo, L. Perrin, W. Pichler, J.-C. Piffaretti, M. Rickenbach (Head of Coordination and Data Center), C. Rudin, P. Sudre, J. Schupbach, A. Telenti, P. Vernazza, R. Weber.