Immune reconstitution pneumonitis following Pneumocystis carinii ...

5 downloads 0 Views 133KB Size Report
An HIV-infected man presented with a pneumonic illness following an episode of treated. Pneumocystis carinii pneumonia (PCP). He had a rise in his CD4 count ...
ß 2002 Blackwell Science Ltd

HIV Medicine (2002), 3, 207±211

CASE REPORT

Immune reconstitution pneumonitis following Pneumocystis carinii pneumonia in HIV-infected subjects SM Barry,1,2 MCI Lipman,1 AR Deery,3 MA Johnson1 and G Janossy2 Departments of 1Thoracic and HIV Medicine, 2Clinical Immunology and 3Pathology, Royal Free Hospital, London, UK

An HIV-infected man presented with a pneumonic illness following an episode of treated Pneumocystis carinii pneumonia (PCP). He had a rise in his CD4 count from 4 to 125 cells/mL on antiretroviral therapy prior to the onset of the second respiratory event. Bronchoalveolar lavage (BAL) revealed no pathogen, although a CD4 lymphocytosis in addition to a highly unusual population of rapidly proliferating CD8 cells was demonstrated. Following 2 weeks of steroid and anti-pneumocystis therapy, a repeat bronchoscopy demonstrated that the expression of these markers had returned to low values. This second respiratory illness, which may have arisen as a consequence of the regenerating immune response reacting to residual P. carinii antigen in the lung, is apparently not rare. When we reviewed our case notes, ®ve further individuals were identi®ed that had started antiretroviral therapy following an episode of PCP and subsequently developed a self-limiting pneumonitis for which no pathogen was identi®ed on bronchoscopy. Keywords: bronchoalveolar lavage, Pneumocystis carinii, pneumonitis Received: 22 March 2002, accepted 14 April 2002 Five further patients were identi®ed that ful®lled the following criteria: (i) cytologically proven PCP treated for 21 days with standard anti-PCP therapy and leading to a complete clinical recovery; (ii) antiretroviral therapy started soon after completion of anti-PCP treatment resulting in a rise in the blood CD4 count; and (iii) a further pneumonic episode for which no pathogen was detected on BAL.

Introduction Highly active antiretroviral therapy (HAART) has been shown to dramatically decrease both HIV-related mortality and morbidity [1]. However, there have been reports of clinical complications associated with the enhanced immune response in patients with Mycobacterium tuberculosis [2, 3], Mycobacterium avium intracellulare [4, 5] and cytomegalovirus infection [6]. The clinical features of the reported cases suggest that a combination of a pre-existing infection together with a vigorous immune response may predispose some patients to secondary in¯ammatory responses. Keeping such immune reconstitution disorders in mind, we identi®ed an HIV-infected patient in whom a pneumonic illness developed following both an episode of Pneumocystis carinii pneumonia (PCP) and effective HAART. No pathogen was identi®ed, but ¯ow cytometry of bronchoalveolar lavage (BAL) demonstrated highly unusual features. These ®ndings suggested that this case could be an immune reconstitution disorder to residual P. carinii antigen. This hypothesis stimulated a case note review to identify further episodes of pneumonitis following PCP infection.

Case reports Patient 1 A 38-year-old HIV-positive man presented with a 2-week history of dyspnoea and a dry cough in August 2000. On examination he was apyrexial, his oxygen saturations on air dropped from 97% at rest to 90% after exercise and his chest radiograph revealed bilateral lower zone shadowing. He was admitted with a provisional diagnosis of PCP which was con®rmed on subsequent BAL. His CD4 count in blood was 4 cells/mL (1%) and his HIV-RNA load was 283 000 copies/ mL. Due to hypersensitivity to sulpha-containing drugs, he was treated with atovaquone and discharged following a good clinical response. Due to the low CD4 count, high HIV viral load and prior antiretroviral history, he was commenced on quadruple therapy with didanosine (ddI), efavirenz, nel®navir and stavudine 5 weeks after completing his antipneumocystis therapy. Two weeks later he developed an illness characterized

Correspondence: Dr Simon Barry, Department of Immunology, Royal Free Hospital, Pond St, London NW3 2QG, UK. Fax: ‡44 207 4310879; e-mail: [email protected]

207

208 SM Barry et al.

by night sweats, cough and then dyspnoea. At the time of this second admission his temperature was 39.4 C and his resting oxygen saturations on air were 91%. A chest radiograph demonstrated bilateral mid and lower zone airspace shadowing. His CD4 count had risen from 4 cells/mL to 125 cells/mL (13%) on the antiretroviral therapy (Table 1) and his HIV-RNA viral load had decreased to 3620 copies/mL. He underwent a further bronchoscopy the day after admission to exclude a recurrence of Pneumocystis carinii pneumonia. No steroid therapy was initiated prior to the bronchoscopy. BAL cytology was negative for P. carinii, acid fast bacilli and viral inclusion bodies. There was no bacterial growth on culture of the lavage and virological examination for respiratory viruses including cytomegalovirus was also negative.

Analysis of BAL and blood was performed immediately after bronchoscopy by ¯ow cytometry. The CD4/CD8 ratio was assessed and the CD8 lymphocytes stained for the presence of the cytotoxic effector molecule perforin and the cell cycling marker Ki67 after permeabilization of the cells. Examination of the BAL revealed a predominance of lymphocytes with a granular appearance (Fig. 1). Twentyseven per cent of these lung lymphocytes were CD4, coinciding with the dramatic rise in blood CD4 lymphocytes. In addition, examination of BAL CD8 lymphocytes demonstrated predominantly proliferative (Ki67‡), perforinpositive cells, that did not express the activation marker CD38 (Fig. 2). On the basis of these highly unusual ®ndings a putative diagnosis of `immune reconstitution pneumonitis' was made and the patient was started on intravenous

Table 1 Blood CD4 counts and severity of pneumonic episodes in HIV-positive cases First episode of proven Pneumocystis carinii pneumonia

Second episode of pneumonitis

Patient

CD4

Severity*

Therapy

CD4

Severity

Duration of antiretroviral therapy{

Therapy

1 2 3 4 5 6

4 70 10 216 290 60

3 4 7 5 5 2

Atovaquone Septrin{ Septrin Septrin steroids Septrin Septrin

125 182 30 340 430 130

6 4 5 5 3 3

19 30 15 5 14 10

Pentamidine steroids Septrin None Septrin steroids Septrin None

* Severity scores as assessed by fever, hypoxia and radiological findings. Maximum score 12 (see Table 2). { Time is number of days on antiretroviral therapy prior to second bronchoalveolar lavage. { Septrin, cotrimoxazole.

× 400 magnification

Fig. 1 Fresh Papanicolau stained bronchoalveolar lavage cytospin from patient 1 at the time of the second episode of pneumonitis, demonstrating a predominance of lymphocytes with a granular appearance (magni®cation  400).

ß 2002 Blackwell Science Ltd HIV Medicine (2002) 3, 207±211

Immune reconstitution pneumonitis following PCP 209

(a)

(b)

128

128 83%

Table 2 Criteria for calculating severity of respiratory episodes

M1

11%

0 0 10

1

2

10

3

10

4

10

10

0 0 10

1

10

Perforin

2

3

10

10

4

10

Perforin

(c)

(d)

128 85%

M1

101

102

103

104

0 0 10

101

Ki 167

(e)

(f)

64

128

0 100

101

102 CD38

102

M1

103

104

Ki 167

1%

69%

M1

103

Score

Fever > 38  C O2 saturation > 95% and desaturation on exertion O2 saturation < 95% > 90% O2 saturation < 90% > 85% O2 saturation < 85% or requiring O2 therapy Chest radiograph shadowing 1 zone Chest radiograph shadowing 2 zones Chest radiograph shadowing 3 zones/generalized Requiring ventiliatory support Death from respiratory episode (maximum possible)

1 1 2 3 4 1 2 3 2 12

128 12%

0 0 10

Feature M1

104

0 100

101

102

103

M1

104

CD38

Fig. 2 Histograms of CD8 lymphocytes from bronchoalveolar lavage depicting the expression of perforin (a,b), Ki67 (c,d) and CD38 (e,f). Histograms a, c and e are from the initial bronchoalveolar lavage; histograms b, d and f are following 2 weeks of steroid therapy. Percentages shown on each histogram relate to the percentage of CD8 lymphocytes that express the relevant marker.

hydrocortisone. Following a cautious clinical approach he was also given intravenous pentamidine. He made a rapid recovery and a chest radiograph 2 weeks later demonstrated considerable improvement. BAL was repeated 2 weeks after starting steroid therapy and again no pneumocystis was seen. The sample was analysed by ¯ow cytometry demonstrating a decline in the CD4 lymphocytes to 6.5% and a profound reduction in perforin and Ki67 expression and an increase in CD38 expression in the CD8 lymphocyte subset (Fig. 2).

Patients 2±6 Five male patients ful®lled the entry criteria. All had their cytologically proven PCP treated with cotrimoxazole (Table 1). Patient 4 was also given concurrent steroids. Anti-pneumocystis therapy resulted in clinical and

ß 2002 Blackwell Science Ltd HIV Medicine (2002) 3, 207±211

O2 saturation ˆ oxygen saturation measured by transcapillary oxymeter.

radiological improvement in all patients. Anti-retroviral therapy (ART) was started a median of 21 (range 17±24) days following the diagnosis of PCP. Patients 2, 3, 4 and 6 were given zidovudine (ZDV), while patient 5 took ZDV and ddI. Following the initiation of ART, all ®ve patients developed respiratory symptoms comprising of fever and dyspnoea, with or without cough and had a bronchoscopy a median of 15 (range 5±30) days after commencing ART. The clinical severity scores for the initial episode of PCP and the second pneumonic episode are displayed in conjunction with CD4 counts before and after ART for all six patients (Table 1). This second respiratory episode was transient and all patients' symptoms had resolved within 1 week. The clinical severity scores were calculated according to a recognized scoring system [7] (Table 2).

Discussion Despite the undeniable clinical bene®ts of HAART, clinicians working in the ®eld of HIV medicine have become aware of the dangers of immune reconstitution illness as patients recover their CD4 counts in the presence of residual antigen [2±6]. Here we describe six cases in whom a pneumonic illness developed following an effectively treated episode of PCP. In all cases BAL preparations were examined by a consultant cytologist for evidence of pneumocystis or mycobacterial infection and for viral cytopathic effects. In addition, the specimens were cultured for bacteria, mycobacteria and fungi and were tested for a range of respiratory viruses. The lack of identi®cation of any pathogenic organisms in BAL in conjunction with a rise in blood CD4 and new respiratory symptoms are suggestive of an immune reconstitution illness. Another possibility could be that the second respiratory illness was in fact a recrudescence of PCP, but this latter eventuality is unlikely for the following reasons. First, the

210 SM Barry et al.

strongest evidence against recurrent PCP and in favour of an immune reconstitution pneumonitis is provided by the detailed analysis of the BAL immunology in patient 1. It is extremely unusual to ®nd such a strong in¯ux of CD4 cells in the lung during HIV infection, resulting in a BAL CD4/ CD8 ratio of 0.54. The comparative median BAL CD4/CD8 ratio in a cohort of 50 HIV-positive patients investigated for a variety of respiratory complaints was only 0.07 (unpublished data). The median blood CD4 count in this cohort was 48 cells/mL. Such an abundance of CD4 lymphocytes in case 1 is likely to have promoted the dramatic wave of proliferation seen in the CD8 cells of the same BAL specimen. These BAL CD8 lymphocytes demonstrated a highly atypical phenotype, 85% of which expressed the cell cycling marker Ki67. In the comparative HIV-positive cohort, the median CD8 expression of Ki67 was only 2.4%. When this HIVpositive cohort was further examined, seven cases of PCP were identi®ed, but the median expression of Ki67 in this subgroup was only 2.3%, thus further supporting our contention that the second pneumonic episode in case 1 was not due to persistent or recurrent PCP. Similarly, perforin was present in 83% of BAL CD8 lymphocytes in patient 1, compared to a median of 2.5% in the HIV cohort. Second, these changes were speci®c for the lung, but not the blood lymphocytes and the changes reversed following steroid therapy. Third, the diagnosis of P. carinii infection was made by an experienced consultant cytologist who examined stained BAL by microscopy. Such a technique is considered to be the `gold standard' against which alternative diagnostic strategies, such as polymerase chain reaction (PCR) techniques, have been evaluated [8, 9]. In particular, since case 1 had a higher severity score for the second pneumonic episode it would have been surprising not to have detected P. carinii if this was really due to a recurrence of the infection. Nevertheless, it is likely that phagocytosed P. carinii antigen is presented by alveolar macrophages to T lymphocytes and that this serves as a powerful immune stimulus in the context of a rapidly regenerating immune response. The pathogenesis of this process and the other putative immune reconstitution disorders attributable to Mycobacteria or cytomegalovirus will be proven when the antigen speci®city of the response is determined. Several immunological techniques have recently been developed to enable antigen speci®c lymphocytes to be enumerated. Both HIV and CMV tetramers with different HLA class 1 subtypes have been used successfully to characterize responses to these infections in the blood [10]. Activation of CD4 lymphocytes by cytokine synthesis [11] and the Elispot method can also be employed. Nevertheless, there is a paucity of published data on the application of these techniques to investigate lung-derived lymphocytes.

Our observations emphasize the value of examining the immune response in the relevant tissues, rather than the blood alone, by demonstrating that the CD8 cell phenotype in the lung varied considerably from that in the blood and altered with steroid therapy, unlike in the blood. Such a clinically orientated analysis of immune responses within the lung may provide both an understanding of the immunopathogenesis and help inform decisions about choosing the relevant therapy in these cases.

Acknowledgements We thank Dr M Tyrer, Ms L Swaden, and Dr SB Squire for their contributions to this manuscript.

References 1 Palella FJ Jr, Delaney KM, Moorman AC et al. Declining morbidity and mortality among patients with advanced human immunode®ciency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338: 853±860. 2 Narita M, Ashkin D, Hollender ES, Pitchenik AE. Paradoxical worsening of tuberculosis following antiretroviral therapy in patients with AIDS. Am J Respir Crit Care Med 1998; 158: 157±161. 3 Crump JA, Tyrer MJ, Lloyd-Owen SJ et al. Military tuberculosis with paradoxical expansion of intracranial tuberculomas complicating human immunode®ciency virus infection in a patient receiving highly active antiretroviral therapy. Clin Infect Dis 1998; 26: 1008±1009. 4 Phillips P, Kwiatkowski MB, Copland M, Craib K, Montaner J. Mycobacterial lymphadenitis associated with the initiation of combination antiretroviral therapy. J Acquir Immune De®c Syndr Hum Retrovirol 1999; 20: 122±128. 5 Race EM, Adelson-Mitty J, Kriegel GR et al. Focal mycobacterial lymphadenitis following initiation of protease-inhibitor therapy in patients with advanced HIV-1 disease. Lancet 1998; 351: 252±255. 6 Zegans ME, Walton RC, Holland GN et al. Transient vitreous in¯ammatory reactions associated with combination antiretroviral therapy in patients with AIDS and cytomegalovirus retinitis. Am J Ophthalmol 1998; 125: 292±300. 7 Squire SB, Lipman MC, Bagdades EK et al. Severe cytomegalovirus pneumonitis in HIV infected patients with higher than average CD4 counts. Thorax 1992; 47: 301±304. 8 Torres J, Goldman M, Wheat LJ et al. Diagnosis of Pneumocystis carinii pneumonia in human immunode®ciency virus-infected patients with polymerase chain reaction: a blinded comparison to standard methods. Clin Infect Dis 2000; 30: 141±145. 9 Huang SN, Fischer SH, O'Shaughnessy E et al. Development of a PCR assay for diagnosis of Pneumocystis carinii pneumonia based

ß 2002 Blackwell Science Ltd HIV Medicine (2002) 3, 207±211

Immune reconstitution pneumonitis following PCP 211

on ampli®cation of the multicopy major surface glycoprotein gene family. Diagn Microbiol Infect Dis 1999; 35: 27±32. 10 Appay V, Nixon DF, Donahoe SM et al. HIV-speci®c CD8 (‡) T cells produce antiviral cytokines but are impaired in cytolytic function. J Exp Med 2000; 192: 63±75.

ß 2002 Blackwell Science Ltd HIV Medicine (2002) 3, 207±211

11 Maino VC, Picker LJ. Identi®cation of functional subsets by ¯ow cytometry: intracellular detection of cytokine expression. Cytometry 1998; 34: 207±215.