- Email: [email protected]
Procalcitonin as a marker of bacterial sepsis in patients infected with HIV-1
Journal of Infection (1997) 35, 41 46
Procalcitonin as a Marker of Bacterial Sepsis in Patients Infected with HIV-1 Y. Gerard 1'2, D. Hober *~, M. Assicot 3, S. AlfandarF, F. Ajana 2, J. M. Bourez 2, C. Chidiac 2, Y. M o u t o n 2, C. Bohuon 3 and P. Wattre 1 1Laboratoire de Virologie, Bdtiment I.R.EEES., CHU, place de Verdun, 59037 Lille Cedex, 2Service Universitaire Rdttional des Maladies Infectieuses et du Voya~eu~ tt@ital Dron, 135 rue du Prdsident Cotg 59208 rourcoing, and 3D@artement de Biologie Clinique, Institut Gustave Roussg 94805 Villejuif, France Procalcitonin (ProCT) is a recently described marker of severe sepsis. It was decided to assess the value of proCT as a marker of secondary infection in patients infected with HIV-1. ProCT plasma levels were measured by immunoluminometric assay in a prospective study in 155 HIV-infected individuals: 102 asymptomatic and 53 with fever or suspected secondary infections. The baseline plasma level of ProCT was low (0.5 n g / m l _ 0.37), even in the latest stages of the disease, and did not differ from the values of healthy subjects (0.54 n g / m l 0.08). EDTA-treated whole blood was collected from patients before starting specific antimicrobial therapy. No elevation of ProCT level was detected in HIV-infected patients with evolving secondary infections including PCP (n = 4), cerebral toxoplasmosis (n =4), viral infections (n = 9), mycobacterial infections (n = 5), localized bacterial (n = 12) and fungal infections (n = 4), malignancies (n = 3), and in various associated infectious and non-infectious febrile events ( n = 13). All these plasma values were lower than 2.1 ng/ml. In contrast, high ProCT plasma levels were detected in one HIV-infected patient with a septicaemic Haemophilus influenzae infection (16.5 ng/ml) and another one with a septicaemic Pseudomonas aeruginosa infection (44.1 ng/ ml). ProCT values decreased rapidly under appropriate therapy. ProCT seems to be a specific marker of bacterial sepsis in HIV-infected patients, as no increase in other secondary infections could be detected in those patients. A rapid determination of ProCT level could be useful to confirm or refute bacterial sepsis for a better management of febrile HIV-infected patients.
Introduction The q u a l i t y of life a n d d u r a t i o n of survival of patients infected w i t h HIV-1 h a s i m p r o v e d d u r i n g the last decade b e c a u s e of a better k n o w l e d g e of t h e infection, c u r a t i v e t r e a t m e n t of o p p o r t u n i s t i c infections, p r e v e n t i v e strategies a n d a n t i r e t r o v i r a l therapy. However, a wide spect r u m of infectious p a t h o g e n s c a n be involved, a n d clinical p r e s e n t a t i o n s are often non-specific. 1 Thus, fever is freq u e n t l y a n isolated s y m p t o m suggestive of infection. The m a i n p r o b l e m e n c o u n t e r e d is in a d a p t i n g the t r e a t m e n t a c c o r d i n g to the p a t h o g e n s involved. Before s t a r t i n g empirical therapy, the p h y s i c i a n m u s t d e t e r m i n e the origin of a n isolated fever: bacterial, parasitic, viral, fungal agents, a n d even m a l i g n a n c i e s c a n be implicated in the o c c u r r e n c e of fever in HIV-infected patients. 1 Therefore a successful m a n a g e m e n t of infections requires biological
assessments a n d n e w m a r k e r s of infection helpful to establish a n a c c u r a t e diagnosis. Recently, h i g h c o n c e n t r a t i o n s of circulating procalcitonin (proCT) h a v e been detected d u r i n g septic conditions in a paediatric u n i t a n d in patients w i t h b u r n injuries in a n intensive care unit, w i t h o u t p r o d u c t i o n of m a t u r e calcitonin. S e r u m proCT c o n c e n t r a t i o n s a p p e a r e d to correlate w i t h the severity of microbial invasion. 2 Previous studies suggested a specific v a l u e of proCT elevation in fever from bacterial origin. Because of i m m u n o s u p p r e s s i o n , HIV-infected patients are h i g h l y susceptible to a great constellation of p a t h o g e n s . Thus it was n e c e s s a r y to assess proCT values a n d to s e a r c h for a specificity of this n e w m a r k e r in different types of s e c o n d a r y infections in these patients.
Patients, Materials and Methods Address correspondence to: Didier Hober Accepted for publication 19 November 1996. This work was presented in part at the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy (San-Francisco, 1995, poster I255).
0163-4453/97/0400414-06 $12.00/0
Patients In a prospective s t u d y c o n d u c t e d b e t w e e n F e b r u a r y a n d July 1 9 9 4 , 155 HIV-infected patients followed in the © 1997 The British Society for the Study of Infection
42
Y. Gdrard et aL
department of infectious diseases in Tourcoing were included. The first group constituted 102 clinically stable HIVinfected patients, randomly selected. They could have had a past history of opportunistic disease. These patients were classified according to 1993 revised CDC classification. 3 Blood samples from this group permitted the determination of proCT values in basal conditions. The second group included 53 patients with symptoms of acute infection. Blood samples were collected for each patient during hospitalization or at admission in the hospital with a suspected diagnosis of infection. Every patient with fever, or unrelated symptoms occurring in the course of HIV-1 infection was routinely enrolled before introduction of any specific treatment. Some of these patients received prophylaxis and/or antiretroviral therapy. Patients gave their consent beforehand. For each patient, at least two sets of blood cultures were taken at different sites. Specific media for isolation of mycobacteria was systematically used for patients with CD4 cell count under lO0/gl. Accurate diagnoses were then established depending on clinical, biological and radiological criteria. Laboratory data were reviewed from patients' charts and some of these data were used for comparisons with proCT values: white blood cell (WBC) counts, renal and hepatic tests, and C-reactive protein (CRP).
Materials Blood was drawn into sterile v a c u u m collection tubes containing EDTA (Vacutainer, Becton, Dickinson U.S.A.). The blood samples were processed within 2 h after collection. Plasma was separated a f t e r centrifugation, and stored at - 20 °C until analysed. Previous studies showed that there was no alteration in proCT dosages with these conditions of storage. 2 A plasma sample containing a high proCT level found in a HIV seronegative patient with a septic shock was collected, stored in the same conditions, then used as a positive control to confirm the reliability of the proCT assay.
Study design In the clinically stable HIV-infected patients, one single measurement of proCT was performed. In order to evaluate the kinetic profile of proCT levels in the second group of patients, several plasma samples were collected in each case of pathological situation: the first one prior to any specific therapy, the others on day 2 and day 8, respectively, after initiating therapy, and the last one w h e n the ~atient was considered cured.
ProCT assay ProCT values were determined with a monoclonal immunoluminometric assay, without knowledge of the patient's clinical data. This assay is adapted from an immunoradiometric assay previously described. 4 The assay is based on the use of two monoclonal antibodies: the first one as a capture antibody directed towards residues 9 6 - 1 0 6 of proCT, and the second one as the tracer antibody directed towards residues 70-76. Capture-antibody coated tubes were incubated with 100 gl of standards or plasma samples, and 100 gl of acridinium ester labelled tracer-antibody (Lumitest procalc, Henning, Berlin, Germany), during 2 h on an orbital shaker ( 3 2 0 r p m ) at room temperature. After three washings with distilled water, an oxidative reaction with hydrogen peroxide generated luminescence, automatically measured in a Berilux analyzer 250 (Behringwerke AG, Marburg, Germany).
Data analysis Clinical and biological data were extracted from patients' files, and analysed using the SAS system (SAS Institute Incorporation, Cary, North Carolina, U.S.A.). Correlations between proCT values and other data were assessed using Pearson's test. The statistical significance of the differences of proCT levels in patients and controls was evaluated by the M a n n - W h i t n e y test.
Results The plasma level of the proCT was evaluated in 155 HIVinfected patients. Mean age was 36.5 years (range 1 5 - 73) and sex ratio 4:12. Risk factors distribution in this study was similar to epidemiological characteristics of the whole cohort followed up in the same unit of infectious diseases, consisting of 51% homo-bisexual men, 31% heterosexuals, 10% intravenous drug abusers, 6% haemophilia or blood recipients, 2% unknown. In 29 healthy subjects, the m e a n plasma level of proCT was 0 . 5 4 n g / m l (range 0-1.66). In order to obtain a better specificity, an upper reference limit was set as superior to 2 ng/ml. The positive control used in the proCT assays elicited a high value: 64 ng/ml. Patients were grouped according to the 1993 CDC classification. The m e a n ( + / - SD) plasma levels of proCT in clinically stable patients at different stages of HIV-1 disease are shown in Table I. Overall, the m e a n value of proCT in plasma of clinically stable patients was 0.5 + / - 0.37 ng/ml (range 0-22), and did not differ statistically from the values observed in healthy subjects.
Bacterial Sepsis in Patients with HIV-1
43
Table I. Baseline plasma ievels of procalcitonin (ng/mI)* in 102 HIV-1 infected patients~ grouped accordirig to the 1993 Revised Classification System. Clinical categories
CD4 + T-cell categories 1 _>500@I 2 200-499/gl 3 <200@I
A Asymptomatic or PGL:)
B Symptomatic not A or C conditions
C AIDS-indicator conditions§
0.32_+0.18 (n=4) 0.46_+0.33 (n=13) 0.52_+0.43 (n=10)
0.54_+0.29 (n~7) 0.45_+0.41 (n=20) 0.63_+0.4:1 (n=22)
0 (n=l) 0.73+0.20 (n=4) 0.44+0.26 (n=21)
* Procalcitonin values were obtained with an immunoluminometric assay. The values are expressed by the mean _+SEM (ng/ml). Healthy controls: O.54 _+0.08 ng/mI. HIV-infected patients presenting with symptoms of acute infection are excluded. $ PGL =persistent generalized lymphadenopathy. § Clinical conditions in category C are listed in the 1993 Revised Classification System. 3
The plasma levels of proCT in 53 patients with symptoms of acute infection are shown in Table lI. High levels of proCT were found in two patients with septicaemia. In patients with parasitic, viral, fungal, mycobacterial infections, bacterial pneumonia or other bacterial localized infections, and febrile lymphoma, the levels of proCT were lower t h a n 2.1 ng/ml. In all the cases of remaining unexplained fevers, no proCT elevation could be documented. In two terminal stage patients, proCT values reached 3.86 and 4.62 ng/ml, respectively. One had gastrointestinal lymphoma, HIV encephalitis and disseminated mycobacterial infection. The other had disseminated mycobacterial infection, staphylococcal pneumonia, Kaposi's sarcoma, and CMV myelitis. These two patients died a few days later. Low serum levels of proCT were confirmed in a retrospective study in patients with PCP, cerebral toxoplasmosis, and mycobacterial infections. However, a moderate elevation was observed in two out of three cases of disseminated infections with Mycobacterium avium complex. No correlations could be found between proCT values and other biological parameters. We studied the variations of proCT levels in HIVinfected patients with septicaemia whose plasma was obtained sequentially during the period indicated. Figure 1 shows the variations of plasma levels of proCT in those patients. The values of proCT were high at the onset of the septicaemia, then returned to the basal level as an appropriate antibiotic therapy was administered soon after the first sample.
Discussion ProCT has never been tested before in HIV-infected patients, to our knowledge. Our study points out a marked
elevation of plasma levels of proCT in bacterial sepsis. In contrast, no elevation could be found in the other cases of infectious and non-infectious manifestations of HIVinduced immunosuppression. Both cases of bacterial sepsis were identified to involve Gram-negative bacteria. Nevertheless, elevated proCT values have been described with Gram-positive bacteria, 3 and the positive control used in the proCT assay was drawn from a patient with a streptococcal sepsis. Our results are in favour of the increase of proCT in disseminated, but not in localized, infections. However, we cannot exclude the possible increase of proCT in HIVinfected patients with disseminated non-bacterial infections, because we reported the elevation of proCT in a HIV seronegative liver transplant recipient with a disseminated candidiasis, s In localized parasitic infections (PCP, toxoplasmosis) we detected normal values of proCT; however, disseminated parasitic infections like acute falciparum malaria can increase the level of proCT, as recently described. 6 With most of the patients, blood samples were collected at the onset of the intercurrent disease. In the group of HIV-infected patients presenting acute symptoms 37 out of 53 consisted of febrile events, with hyperthermia over 38 °C. Some of the patients were receiving antiretroviral therapy or prophylactic medications. We cannot exclude an interference between these molecules and proCT in our assay, which could account for low levels of proCT. However, patients with opportunistic infections as a first indicator of HIV-infection, free of any treatment, had normal proCT values. In some patients with long-standing fevers, the underlying cause could not be determined despite careful examination, chest X-ray films, blood tests and cultures. Empirical therapy was started before the diagnosis was carried out. Sometimes the diagnosis of a mycobacterial
44
K G4rard e t al, Table II. Procalcitonin values in HIV-infected patients with secondary infections.
Bacterial infections Pseudomonas aeruginosa septicaemia Haemophilus influenzae septicaemia Streptococcus pneumoniae p n e u m o n i a Staphylococcal p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a / l u n g carcinoma Dental infection Sore throat Staphylococcal sinusitis Streptococcal skin cellulitis Pseudomonas colitis Listeria encephalitis
Temperature ( ° C )
Procalcitonin* (ng/mI)
CRP~ (rag~l)
41 40 37.4 38 40 38,3 38.7 39.3 39.5 39.3 39 37 38.9 40
16.46 44.12 1.79 0.69 0.28 0.54 0.22 1.92 1.66 1.3 0.07 0.45 0,15 2,1
179 1 132 2 24 nd ~ nd 266 15 4 nd nd 21 87
-38 40
0.05 0.34 0.73
42 116 80
38,3
O.15
39
0
nd
38.2 39 38.5 36.5 37 37.5 39.6
0.29 0.14 0.58 0.63 0.29 0.08 0.51
55 39 nd 7 nd nd 2
37 37 38.5 39.7 37 37 37 39.6 37.3
1.48 0.1 0.27 1.79 O. 89 0.05 0.05 0.51 1.57
nd nd nd
37 38.8
0 0.24
nd
37 39 39 39 39 39 40.3 38 39 38.8 37.5 37.8 38 38.5 38.2 39 37.2
0.51 0.87 0.38 0.55 1.04 0.44 0.81 0.13 0.53 O. 88 1.41 0.91 0.84 0.17 0.76 3.86 4.62
nd 36 nd 1 0 3 180 nd 59 3 4 11 1 0 2 170 54
Mycobacterial infections Pulmonary tuberculosis Pulmonary tuberculosis Disseminated tuberculosis Disseminated M. avium Disseminated M, avium
Parasitic infections PCP (Pneumocystic carinii pneumonia) PCP PCP PCP + cerebral toxoplasmosis Cerebral toxoplasmosis Cerebral toxoplasmosis Recurrent cerebral toxoplasmosis
5
Viral infections Active chronic viral B hepatitis Acute viral B hepatitis HIV encephalitis + unexplained fever Acute HIV seroconversion Rhinitis CMV retinitis CMV retinitis Recurrent CMV retinitts Progressive multifocal leucoencephalopathy
Fungal infections Candida oesophagitis Aspergillus and Candida p n e u m o n i a
1 nd nd nd 2 8
0
Miscellaneous KS§ Cerebral l y m p h o m a KS + l y m p h o m a + CMV colitis Candida oesophagitis + cryptosporidiosis + sinusitis Candida oesophagitis + sinusitis Disseminated M, avium + HIV myelitis PCP + pulmonary tuberculosis Pancreatitis Unexplained fever+ oral aphthous ulcers Unexplained fever + myocardiopathy Unexplained fever + HIV encephalitis Unexplained fever Unexplained fever Unexplained fever Drug fever Terminal stage: l y m p h o m a + M. avium Terminal stage: KS + CMV myelitis + MycobacteriaI avium + bronchopneumonia
* Procalcitonin plasma levels were determined with an immunoluminometric assay. Data s h o w n represent values before initiating therapy (except for recurrent diseases). Normal value in healthy subjects: <2 ng/ml. I CRP = C reactive protein. Normal value <8 mg/1. :~n d = not done. § KS =Kaposi's sarcoma.
Bacterial Sepsis in Patients with HIV-1
45
was interesting to measure proCT levels in such patients to determine t h e potential value of this marker in MAC | infection. However, it was not possible to detect an 4O increase of proCT level in mycobacterial infections during this prospective study, despite blood and organ biopsies a0 = positive cultures. E '~ 20 This study confirms that proCT levels are high in bacterial sepsis. Moreover, kinetic profiles of proCT could ~ 10 be useful in the follow-up of a sepsis under therapy. Indeed, the finding of a high proCT level in plasma of I I febrile HIV-infected subjects is in favour of the diagnosis I 4 5 of septicaemia and argues for the immediate starting of Days post onset of infection therapy. By contrast, no elevation could be found in other secondary infectious diseases in patients. Bacterial 60 infections, although less common than opportunistic infections among HIV-infected patients, cause a significant 50-amount of morbidity and mortality. 1° Indeed, predisposing -~ 40 factors include alterations in skin and mucous membranes. B-cell dysfunction, a disordered phagocyte func80 'Ca tion, granulocytopenia resulting from use of antimicrobial ~ 20q and cytotoxic agents, and a high risk of infection of intravenous catheters. High incidence of bacteraemia in ~ 10 these patients, with the occurrence of fatal cases, was I reported, n'~2 The number of septicaemic patients was 0 2 16 small in this prospective study, but remained repDays post onset of infection resentative of the general incidence of communityFigure 1. Follow-up of plasma levels of procalcitonin in two HIV-1 acquired bacteraemia in HIV-infected patients, which infected patients with septicaemia. EDTA-treated whole blood was represents approximately 5 % of the admissions of patients collected at the time of admission (day 0), just before starting appropriate w i t h AIDS. it Because both septicaemic patients had high antibacterial therapy. Procalcitonin levels were obtained by an improCT values, this marker may exhibit a high specificity munoluminometric assay. Normal values <2 ng/ml. Diagnosis was proven with positive blood cultures. (a) Haemophilus influenzae sepfor bacterial sepsis. Other markers of infection, like cticaemia. (b) Pseudomonas aeruginosa septicaemia. reactive protein (CRP) measurements, were not systematically performed. Nevertheless, Table II shows very different CRP values for each category of infection, suginfection was established a few days or a few weeks later. gesting the absence of good specificity. Further obIn two patients in the terminal phase of their illness, servations are warranted to confirm proCT usefulness as we obtained a slight elevation of proCT. Organ failures an indicator of serious bacterial infection, for a better such as liver deficiency and renal dysfunction could raise management of febrile HIV-infected patients. the concentration of proCT in peripheral blood. However, The cellular source and biological mechanisms of proCT the proCT levels in those patients were much lower production still remain unclear. However, recently it has than the ones in patients with septicaemia. Nevertheless, been shown that endotoxin induces the release of proCT elevated proCT values could argue for a poor prognosis, in normal subjects, which is related to elevations of tumor as both patients died a few days later. The same ob- necrosis factor-a and interleukin-6. ~3 Further studies are servation of poor prognosis for elevated proCT values required to investigate the regulation of proCT prowas found in acute melioidosis, an infection caused by duction, its potential role in the pathogenesis of infectious Pseudomonas pseudornallei and characterized by a wide diseases and its value as a marker of infection. range of clinical manifestations.7 Mycobacterium avium complex (MAC) is responsible for References a high grade mycobacteremia affecting multiple organs J_ Masur H. Management of opportunistic infections associated with and is widely disseminated in AIDS patients, s Clinical HIV infection. In: Mandell GL, Douglas RG, Bennett JE, eds. Principles signs and symptoms are non-specific, and therefore difal~d practice of infectious diseases 4th ed. New York: Churchill Livficult to distinguish from other AIDS-related conditions, ingstone. 1995; 1280-1294. 2 Assicot M, Gendrel D, Carsin H, Raymond I, Guilband J, Bohuon requiring positive mycobacterial cultures. 9 Therefore it 501
4.a
46
3
4
5
6
7
8
Y. Gdrard et al, C. High serum procalcitonin concentrations in patients with sepsis infection. Lancet 1993; 341: 515-518. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. M M W R 1992 41: 1-19. Ghillani P, Motte P, Troalen F et al. Identification and measurement of calcitonin precursors in serum of patients with malignant diseases, Cancer Res 1989; 49: 6845-6851. G~rard Y, Hober D, Petijean Set al. High serum procalcitonin level in a 4-year-old liver transplant recipient with a disseminated candidiasis. Infection 1995; 23: 310-311. Davis TME, Assicot M, Bohuon C, St John A, Li GQ, Anh TIC Serum procalcitonin concentrations in acute malaria. 7?rans R Soc Trop Med Hyg 1994; 88: 670-671. Smith MD, Suputtamongkol Y, Chaowagul Wet al. Elevated serum procalcitonin levels in patients with meliodosis. Clin Infect Dis 1995; 20: 641-645. Benson CA, Ellner lJ. Mycobacterium avium Complex infection and AIDS: advances in theory and practice. Clin Infect Dis 1993; 17: 7-20.
9 Pitchenik AE, FerteI D. Mycobacterial disease in patients with HIV infection. In: Wormser GP, ed. AIDS and other manifestations of HIV infection. 2nd ed. New York: Raven Press. 1992: 297-313. 10 Simberkoff MS, Leaf HL. Bacterial infections in patients with HIV infection. In: Wormser GP, ed. AIDS and other manifestations of IIIV infection. 2nd ed. New York: Raven Press. 1992: 269-275. 11 Krumholtz HM, Sande MA, Lo B. Community-acquired bacteremia in patients with acquired immunodeficiency syndrome: clinical presentation, bacteriology, and outcome. Am ] Meal 1989; 86: 776-770. 12 Chidiac C, Alfandari S, Sennevifie E et al. Prospective study of bacterial infections in patients with HIV infection in the north France AIDS reference center (CISII-I 18). In: Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA, U.S.A. American Society for Microbiology 1995: 251. 13 Dandona P, Nix D, Wilson MF et aI. Procalcitonin increase after endotoxin injection in normal subjects. ] CIin Endoerinol Metab 199zi; 79: 1605-1608.
Procalcitonin as a Marker of Bacterial Sepsis in Patients Infected with HIV-1 Y. Gerard 1'2, D. Hober *~, M. Assicot 3, S. AlfandarF, F. Ajana 2, J. M. Bourez 2, C. Chidiac 2, Y. M o u t o n 2, C. Bohuon 3 and P. Wattre 1 1Laboratoire de Virologie, Bdtiment I.R.EEES., CHU, place de Verdun, 59037 Lille Cedex, 2Service Universitaire Rdttional des Maladies Infectieuses et du Voya~eu~ tt@ital Dron, 135 rue du Prdsident Cotg 59208 rourcoing, and 3D@artement de Biologie Clinique, Institut Gustave Roussg 94805 Villejuif, France Procalcitonin (ProCT) is a recently described marker of severe sepsis. It was decided to assess the value of proCT as a marker of secondary infection in patients infected with HIV-1. ProCT plasma levels were measured by immunoluminometric assay in a prospective study in 155 HIV-infected individuals: 102 asymptomatic and 53 with fever or suspected secondary infections. The baseline plasma level of ProCT was low (0.5 n g / m l _ 0.37), even in the latest stages of the disease, and did not differ from the values of healthy subjects (0.54 n g / m l 0.08). EDTA-treated whole blood was collected from patients before starting specific antimicrobial therapy. No elevation of ProCT level was detected in HIV-infected patients with evolving secondary infections including PCP (n = 4), cerebral toxoplasmosis (n =4), viral infections (n = 9), mycobacterial infections (n = 5), localized bacterial (n = 12) and fungal infections (n = 4), malignancies (n = 3), and in various associated infectious and non-infectious febrile events ( n = 13). All these plasma values were lower than 2.1 ng/ml. In contrast, high ProCT plasma levels were detected in one HIV-infected patient with a septicaemic Haemophilus influenzae infection (16.5 ng/ml) and another one with a septicaemic Pseudomonas aeruginosa infection (44.1 ng/ ml). ProCT values decreased rapidly under appropriate therapy. ProCT seems to be a specific marker of bacterial sepsis in HIV-infected patients, as no increase in other secondary infections could be detected in those patients. A rapid determination of ProCT level could be useful to confirm or refute bacterial sepsis for a better management of febrile HIV-infected patients.
Introduction The q u a l i t y of life a n d d u r a t i o n of survival of patients infected w i t h HIV-1 h a s i m p r o v e d d u r i n g the last decade b e c a u s e of a better k n o w l e d g e of t h e infection, c u r a t i v e t r e a t m e n t of o p p o r t u n i s t i c infections, p r e v e n t i v e strategies a n d a n t i r e t r o v i r a l therapy. However, a wide spect r u m of infectious p a t h o g e n s c a n be involved, a n d clinical p r e s e n t a t i o n s are often non-specific. 1 Thus, fever is freq u e n t l y a n isolated s y m p t o m suggestive of infection. The m a i n p r o b l e m e n c o u n t e r e d is in a d a p t i n g the t r e a t m e n t a c c o r d i n g to the p a t h o g e n s involved. Before s t a r t i n g empirical therapy, the p h y s i c i a n m u s t d e t e r m i n e the origin of a n isolated fever: bacterial, parasitic, viral, fungal agents, a n d even m a l i g n a n c i e s c a n be implicated in the o c c u r r e n c e of fever in HIV-infected patients. 1 Therefore a successful m a n a g e m e n t of infections requires biological
assessments a n d n e w m a r k e r s of infection helpful to establish a n a c c u r a t e diagnosis. Recently, h i g h c o n c e n t r a t i o n s of circulating procalcitonin (proCT) h a v e been detected d u r i n g septic conditions in a paediatric u n i t a n d in patients w i t h b u r n injuries in a n intensive care unit, w i t h o u t p r o d u c t i o n of m a t u r e calcitonin. S e r u m proCT c o n c e n t r a t i o n s a p p e a r e d to correlate w i t h the severity of microbial invasion. 2 Previous studies suggested a specific v a l u e of proCT elevation in fever from bacterial origin. Because of i m m u n o s u p p r e s s i o n , HIV-infected patients are h i g h l y susceptible to a great constellation of p a t h o g e n s . Thus it was n e c e s s a r y to assess proCT values a n d to s e a r c h for a specificity of this n e w m a r k e r in different types of s e c o n d a r y infections in these patients.
Patients, Materials and Methods Address correspondence to: Didier Hober Accepted for publication 19 November 1996. This work was presented in part at the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy (San-Francisco, 1995, poster I255).
0163-4453/97/0400414-06 $12.00/0
Patients In a prospective s t u d y c o n d u c t e d b e t w e e n F e b r u a r y a n d July 1 9 9 4 , 155 HIV-infected patients followed in the © 1997 The British Society for the Study of Infection
42
Y. Gdrard et aL
department of infectious diseases in Tourcoing were included. The first group constituted 102 clinically stable HIVinfected patients, randomly selected. They could have had a past history of opportunistic disease. These patients were classified according to 1993 revised CDC classification. 3 Blood samples from this group permitted the determination of proCT values in basal conditions. The second group included 53 patients with symptoms of acute infection. Blood samples were collected for each patient during hospitalization or at admission in the hospital with a suspected diagnosis of infection. Every patient with fever, or unrelated symptoms occurring in the course of HIV-1 infection was routinely enrolled before introduction of any specific treatment. Some of these patients received prophylaxis and/or antiretroviral therapy. Patients gave their consent beforehand. For each patient, at least two sets of blood cultures were taken at different sites. Specific media for isolation of mycobacteria was systematically used for patients with CD4 cell count under lO0/gl. Accurate diagnoses were then established depending on clinical, biological and radiological criteria. Laboratory data were reviewed from patients' charts and some of these data were used for comparisons with proCT values: white blood cell (WBC) counts, renal and hepatic tests, and C-reactive protein (CRP).
Materials Blood was drawn into sterile v a c u u m collection tubes containing EDTA (Vacutainer, Becton, Dickinson U.S.A.). The blood samples were processed within 2 h after collection. Plasma was separated a f t e r centrifugation, and stored at - 20 °C until analysed. Previous studies showed that there was no alteration in proCT dosages with these conditions of storage. 2 A plasma sample containing a high proCT level found in a HIV seronegative patient with a septic shock was collected, stored in the same conditions, then used as a positive control to confirm the reliability of the proCT assay.
Study design In the clinically stable HIV-infected patients, one single measurement of proCT was performed. In order to evaluate the kinetic profile of proCT levels in the second group of patients, several plasma samples were collected in each case of pathological situation: the first one prior to any specific therapy, the others on day 2 and day 8, respectively, after initiating therapy, and the last one w h e n the ~atient was considered cured.
ProCT assay ProCT values were determined with a monoclonal immunoluminometric assay, without knowledge of the patient's clinical data. This assay is adapted from an immunoradiometric assay previously described. 4 The assay is based on the use of two monoclonal antibodies: the first one as a capture antibody directed towards residues 9 6 - 1 0 6 of proCT, and the second one as the tracer antibody directed towards residues 70-76. Capture-antibody coated tubes were incubated with 100 gl of standards or plasma samples, and 100 gl of acridinium ester labelled tracer-antibody (Lumitest procalc, Henning, Berlin, Germany), during 2 h on an orbital shaker ( 3 2 0 r p m ) at room temperature. After three washings with distilled water, an oxidative reaction with hydrogen peroxide generated luminescence, automatically measured in a Berilux analyzer 250 (Behringwerke AG, Marburg, Germany).
Data analysis Clinical and biological data were extracted from patients' files, and analysed using the SAS system (SAS Institute Incorporation, Cary, North Carolina, U.S.A.). Correlations between proCT values and other data were assessed using Pearson's test. The statistical significance of the differences of proCT levels in patients and controls was evaluated by the M a n n - W h i t n e y test.
Results The plasma level of the proCT was evaluated in 155 HIVinfected patients. Mean age was 36.5 years (range 1 5 - 73) and sex ratio 4:12. Risk factors distribution in this study was similar to epidemiological characteristics of the whole cohort followed up in the same unit of infectious diseases, consisting of 51% homo-bisexual men, 31% heterosexuals, 10% intravenous drug abusers, 6% haemophilia or blood recipients, 2% unknown. In 29 healthy subjects, the m e a n plasma level of proCT was 0 . 5 4 n g / m l (range 0-1.66). In order to obtain a better specificity, an upper reference limit was set as superior to 2 ng/ml. The positive control used in the proCT assays elicited a high value: 64 ng/ml. Patients were grouped according to the 1993 CDC classification. The m e a n ( + / - SD) plasma levels of proCT in clinically stable patients at different stages of HIV-1 disease are shown in Table I. Overall, the m e a n value of proCT in plasma of clinically stable patients was 0.5 + / - 0.37 ng/ml (range 0-22), and did not differ statistically from the values observed in healthy subjects.
Bacterial Sepsis in Patients with HIV-1
43
Table I. Baseline plasma ievels of procalcitonin (ng/mI)* in 102 HIV-1 infected patients~ grouped accordirig to the 1993 Revised Classification System. Clinical categories
CD4 + T-cell categories 1 _>500@I 2 200-499/gl 3 <200@I
A Asymptomatic or PGL:)
B Symptomatic not A or C conditions
C AIDS-indicator conditions§
0.32_+0.18 (n=4) 0.46_+0.33 (n=13) 0.52_+0.43 (n=10)
0.54_+0.29 (n~7) 0.45_+0.41 (n=20) 0.63_+0.4:1 (n=22)
0 (n=l) 0.73+0.20 (n=4) 0.44+0.26 (n=21)
* Procalcitonin values were obtained with an immunoluminometric assay. The values are expressed by the mean _+SEM (ng/ml). Healthy controls: O.54 _+0.08 ng/mI. HIV-infected patients presenting with symptoms of acute infection are excluded. $ PGL =persistent generalized lymphadenopathy. § Clinical conditions in category C are listed in the 1993 Revised Classification System. 3
The plasma levels of proCT in 53 patients with symptoms of acute infection are shown in Table lI. High levels of proCT were found in two patients with septicaemia. In patients with parasitic, viral, fungal, mycobacterial infections, bacterial pneumonia or other bacterial localized infections, and febrile lymphoma, the levels of proCT were lower t h a n 2.1 ng/ml. In all the cases of remaining unexplained fevers, no proCT elevation could be documented. In two terminal stage patients, proCT values reached 3.86 and 4.62 ng/ml, respectively. One had gastrointestinal lymphoma, HIV encephalitis and disseminated mycobacterial infection. The other had disseminated mycobacterial infection, staphylococcal pneumonia, Kaposi's sarcoma, and CMV myelitis. These two patients died a few days later. Low serum levels of proCT were confirmed in a retrospective study in patients with PCP, cerebral toxoplasmosis, and mycobacterial infections. However, a moderate elevation was observed in two out of three cases of disseminated infections with Mycobacterium avium complex. No correlations could be found between proCT values and other biological parameters. We studied the variations of proCT levels in HIVinfected patients with septicaemia whose plasma was obtained sequentially during the period indicated. Figure 1 shows the variations of plasma levels of proCT in those patients. The values of proCT were high at the onset of the septicaemia, then returned to the basal level as an appropriate antibiotic therapy was administered soon after the first sample.
Discussion ProCT has never been tested before in HIV-infected patients, to our knowledge. Our study points out a marked
elevation of plasma levels of proCT in bacterial sepsis. In contrast, no elevation could be found in the other cases of infectious and non-infectious manifestations of HIVinduced immunosuppression. Both cases of bacterial sepsis were identified to involve Gram-negative bacteria. Nevertheless, elevated proCT values have been described with Gram-positive bacteria, 3 and the positive control used in the proCT assay was drawn from a patient with a streptococcal sepsis. Our results are in favour of the increase of proCT in disseminated, but not in localized, infections. However, we cannot exclude the possible increase of proCT in HIVinfected patients with disseminated non-bacterial infections, because we reported the elevation of proCT in a HIV seronegative liver transplant recipient with a disseminated candidiasis, s In localized parasitic infections (PCP, toxoplasmosis) we detected normal values of proCT; however, disseminated parasitic infections like acute falciparum malaria can increase the level of proCT, as recently described. 6 With most of the patients, blood samples were collected at the onset of the intercurrent disease. In the group of HIV-infected patients presenting acute symptoms 37 out of 53 consisted of febrile events, with hyperthermia over 38 °C. Some of the patients were receiving antiretroviral therapy or prophylactic medications. We cannot exclude an interference between these molecules and proCT in our assay, which could account for low levels of proCT. However, patients with opportunistic infections as a first indicator of HIV-infection, free of any treatment, had normal proCT values. In some patients with long-standing fevers, the underlying cause could not be determined despite careful examination, chest X-ray films, blood tests and cultures. Empirical therapy was started before the diagnosis was carried out. Sometimes the diagnosis of a mycobacterial
44
K G4rard e t al, Table II. Procalcitonin values in HIV-infected patients with secondary infections.
Bacterial infections Pseudomonas aeruginosa septicaemia Haemophilus influenzae septicaemia Streptococcus pneumoniae p n e u m o n i a Staphylococcal p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a Bacterial p n e u m o n i a / l u n g carcinoma Dental infection Sore throat Staphylococcal sinusitis Streptococcal skin cellulitis Pseudomonas colitis Listeria encephalitis
Temperature ( ° C )
Procalcitonin* (ng/mI)
CRP~ (rag~l)
41 40 37.4 38 40 38,3 38.7 39.3 39.5 39.3 39 37 38.9 40
16.46 44.12 1.79 0.69 0.28 0.54 0.22 1.92 1.66 1.3 0.07 0.45 0,15 2,1
179 1 132 2 24 nd ~ nd 266 15 4 nd nd 21 87
-38 40
0.05 0.34 0.73
42 116 80
38,3
O.15
39
0
nd
38.2 39 38.5 36.5 37 37.5 39.6
0.29 0.14 0.58 0.63 0.29 0.08 0.51
55 39 nd 7 nd nd 2
37 37 38.5 39.7 37 37 37 39.6 37.3
1.48 0.1 0.27 1.79 O. 89 0.05 0.05 0.51 1.57
nd nd nd
37 38.8
0 0.24
nd
37 39 39 39 39 39 40.3 38 39 38.8 37.5 37.8 38 38.5 38.2 39 37.2
0.51 0.87 0.38 0.55 1.04 0.44 0.81 0.13 0.53 O. 88 1.41 0.91 0.84 0.17 0.76 3.86 4.62
nd 36 nd 1 0 3 180 nd 59 3 4 11 1 0 2 170 54
Mycobacterial infections Pulmonary tuberculosis Pulmonary tuberculosis Disseminated tuberculosis Disseminated M. avium Disseminated M, avium
Parasitic infections PCP (Pneumocystic carinii pneumonia) PCP PCP PCP + cerebral toxoplasmosis Cerebral toxoplasmosis Cerebral toxoplasmosis Recurrent cerebral toxoplasmosis
5
Viral infections Active chronic viral B hepatitis Acute viral B hepatitis HIV encephalitis + unexplained fever Acute HIV seroconversion Rhinitis CMV retinitis CMV retinitis Recurrent CMV retinitts Progressive multifocal leucoencephalopathy
Fungal infections Candida oesophagitis Aspergillus and Candida p n e u m o n i a
1 nd nd nd 2 8
0
Miscellaneous KS§ Cerebral l y m p h o m a KS + l y m p h o m a + CMV colitis Candida oesophagitis + cryptosporidiosis + sinusitis Candida oesophagitis + sinusitis Disseminated M, avium + HIV myelitis PCP + pulmonary tuberculosis Pancreatitis Unexplained fever+ oral aphthous ulcers Unexplained fever + myocardiopathy Unexplained fever + HIV encephalitis Unexplained fever Unexplained fever Unexplained fever Drug fever Terminal stage: l y m p h o m a + M. avium Terminal stage: KS + CMV myelitis + MycobacteriaI avium + bronchopneumonia
* Procalcitonin plasma levels were determined with an immunoluminometric assay. Data s h o w n represent values before initiating therapy (except for recurrent diseases). Normal value in healthy subjects: <2 ng/ml. I CRP = C reactive protein. Normal value <8 mg/1. :~n d = not done. § KS =Kaposi's sarcoma.
Bacterial Sepsis in Patients with HIV-1
45
was interesting to measure proCT levels in such patients to determine t h e potential value of this marker in MAC | infection. However, it was not possible to detect an 4O increase of proCT level in mycobacterial infections during this prospective study, despite blood and organ biopsies a0 = positive cultures. E '~ 20 This study confirms that proCT levels are high in bacterial sepsis. Moreover, kinetic profiles of proCT could ~ 10 be useful in the follow-up of a sepsis under therapy. Indeed, the finding of a high proCT level in plasma of I I febrile HIV-infected subjects is in favour of the diagnosis I 4 5 of septicaemia and argues for the immediate starting of Days post onset of infection therapy. By contrast, no elevation could be found in other secondary infectious diseases in patients. Bacterial 60 infections, although less common than opportunistic infections among HIV-infected patients, cause a significant 50-amount of morbidity and mortality. 1° Indeed, predisposing -~ 40 factors include alterations in skin and mucous membranes. B-cell dysfunction, a disordered phagocyte func80 'Ca tion, granulocytopenia resulting from use of antimicrobial ~ 20q and cytotoxic agents, and a high risk of infection of intravenous catheters. High incidence of bacteraemia in ~ 10 these patients, with the occurrence of fatal cases, was I reported, n'~2 The number of septicaemic patients was 0 2 16 small in this prospective study, but remained repDays post onset of infection resentative of the general incidence of communityFigure 1. Follow-up of plasma levels of procalcitonin in two HIV-1 acquired bacteraemia in HIV-infected patients, which infected patients with septicaemia. EDTA-treated whole blood was represents approximately 5 % of the admissions of patients collected at the time of admission (day 0), just before starting appropriate w i t h AIDS. it Because both septicaemic patients had high antibacterial therapy. Procalcitonin levels were obtained by an improCT values, this marker may exhibit a high specificity munoluminometric assay. Normal values <2 ng/ml. Diagnosis was proven with positive blood cultures. (a) Haemophilus influenzae sepfor bacterial sepsis. Other markers of infection, like cticaemia. (b) Pseudomonas aeruginosa septicaemia. reactive protein (CRP) measurements, were not systematically performed. Nevertheless, Table II shows very different CRP values for each category of infection, suginfection was established a few days or a few weeks later. gesting the absence of good specificity. Further obIn two patients in the terminal phase of their illness, servations are warranted to confirm proCT usefulness as we obtained a slight elevation of proCT. Organ failures an indicator of serious bacterial infection, for a better such as liver deficiency and renal dysfunction could raise management of febrile HIV-infected patients. the concentration of proCT in peripheral blood. However, The cellular source and biological mechanisms of proCT the proCT levels in those patients were much lower production still remain unclear. However, recently it has than the ones in patients with septicaemia. Nevertheless, been shown that endotoxin induces the release of proCT elevated proCT values could argue for a poor prognosis, in normal subjects, which is related to elevations of tumor as both patients died a few days later. The same ob- necrosis factor-a and interleukin-6. ~3 Further studies are servation of poor prognosis for elevated proCT values required to investigate the regulation of proCT prowas found in acute melioidosis, an infection caused by duction, its potential role in the pathogenesis of infectious Pseudomonas pseudornallei and characterized by a wide diseases and its value as a marker of infection. range of clinical manifestations.7 Mycobacterium avium complex (MAC) is responsible for References a high grade mycobacteremia affecting multiple organs J_ Masur H. Management of opportunistic infections associated with and is widely disseminated in AIDS patients, s Clinical HIV infection. In: Mandell GL, Douglas RG, Bennett JE, eds. Principles signs and symptoms are non-specific, and therefore difal~d practice of infectious diseases 4th ed. New York: Churchill Livficult to distinguish from other AIDS-related conditions, ingstone. 1995; 1280-1294. 2 Assicot M, Gendrel D, Carsin H, Raymond I, Guilband J, Bohuon requiring positive mycobacterial cultures. 9 Therefore it 501
4.a
46
3
4
5
6
7
8
Y. Gdrard et al, C. High serum procalcitonin concentrations in patients with sepsis infection. Lancet 1993; 341: 515-518. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. M M W R 1992 41: 1-19. Ghillani P, Motte P, Troalen F et al. Identification and measurement of calcitonin precursors in serum of patients with malignant diseases, Cancer Res 1989; 49: 6845-6851. G~rard Y, Hober D, Petijean Set al. High serum procalcitonin level in a 4-year-old liver transplant recipient with a disseminated candidiasis. Infection 1995; 23: 310-311. Davis TME, Assicot M, Bohuon C, St John A, Li GQ, Anh TIC Serum procalcitonin concentrations in acute malaria. 7?rans R Soc Trop Med Hyg 1994; 88: 670-671. Smith MD, Suputtamongkol Y, Chaowagul Wet al. Elevated serum procalcitonin levels in patients with meliodosis. Clin Infect Dis 1995; 20: 641-645. Benson CA, Ellner lJ. Mycobacterium avium Complex infection and AIDS: advances in theory and practice. Clin Infect Dis 1993; 17: 7-20.
9 Pitchenik AE, FerteI D. Mycobacterial disease in patients with HIV infection. In: Wormser GP, ed. AIDS and other manifestations of HIV infection. 2nd ed. New York: Raven Press. 1992: 297-313. 10 Simberkoff MS, Leaf HL. Bacterial infections in patients with HIV infection. In: Wormser GP, ed. AIDS and other manifestations of IIIV infection. 2nd ed. New York: Raven Press. 1992: 269-275. 11 Krumholtz HM, Sande MA, Lo B. Community-acquired bacteremia in patients with acquired immunodeficiency syndrome: clinical presentation, bacteriology, and outcome. Am ] Meal 1989; 86: 776-770. 12 Chidiac C, Alfandari S, Sennevifie E et al. Prospective study of bacterial infections in patients with HIV infection in the north France AIDS reference center (CISII-I 18). In: Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA, U.S.A. American Society for Microbiology 1995: 251. 13 Dandona P, Nix D, Wilson MF et aI. Procalcitonin increase after endotoxin injection in normal subjects. ] CIin Endoerinol Metab 199zi; 79: 1605-1608.