[Suicidal ideas in psychiatric emergency departments: prospective study comparing self- and hetero-assessment]. / Idées suicidaires aux urgences psychiatriques : étude prospective comparant auto- et hétéro-évaluation.

OBJECTIVE: Many suicide victims had contacts with an emergency department before their attempt. We aimed to determine whether patients coming to a psychiatric emergency department were well assessed concerning their suicidal risk, and to test an easy to fill in scale rapidly assessing suicidal risk. METHOD: We conducted a descriptive epidemiological survey in Marseille. The source population was all patients admitted to the psychiatric emergency department. We used a booklet containing three questionnaires for "nurse", "psychiatrist" and "patient". We estimated the suicidal risk using both a visual analogue scale (similar for patients and caregivers), and validated scales on self-assessment (scale of suicidality SBQ-R and the Beck Hopelessness Scale). RESULTS AND DISCUSSION: The questionnaire results have shown that people who visited a psychiatric emergency department presented a significant suicidal risk on several criteria: socio-demographic criteria (social isolation, low level of education, low number of people with a job), psychiatric history (rate of pre-existing psychiatric disorders significantly higher than in the general population, high proportions of family and personal history of suicide attempts, psychiatric hospitalizations, and people with a psychiatrist). Six percent of patients claimed to have come to an emergency unit for suicidal ideas but they were ten times more with a suicidal risk, according to the SBQ-R score. The suicidal risk self-assessed by patients on our visual analogue scale was well correlated with SBQ-R scale and Beck Hopelessness scale, but was not well correlated with the evaluation of caregivers. CONCLUSION: Hence, the analog scale we created is easy to use and seems to be a good tool for suicidal risk estimation when it is self-assessed by patients in our study population.

A single dose of ciprofloxacin reduces the duration of diarrhea among service members deployed in Africa.

OBJECTIVES: To assess the effectiveness of the adjunction of a one-gram single dose of ciprofloxacin to a symptomatic treatment for the early treatment of uncomplicated diarrhea during military operations of the French service members in Africa. PATIENTS AND METHODS: This phase IV, multicentric, randomized, open-label, controlled trial was conducted in Chad, Mali, and in Central African Republic. A total of 267 French service members having at least one loose stool in the previous 24 hours were enrolled from May 2015 to June 2016. Participants were randomized to receive ciprofloxacin 1 g and a symptomatic treatment (racecadotril 100 mg three times a day and ad libidum oral rehydration solution) or a symptomatic treatment alone. The primary outcome was the duration of the diarrhea. Secondary outcomes were evaluated at the 72-hour endpoint and included recovery status, number of loose stools, frequency and duration of associated symptoms and safety of treatments. RESULTS: Among 267 participants, 242 completed the trial. Participants receiving ciprofloxacin and a symptomatic treatment (n = 124) were significantly more likely to be cured at the endpoint than those who only received a symptomatic treatment (118): 94.4 % versus 74.6 % (OR = 5.7; 95 %CI: [2.4-13.6]; p < 10-3). The antibiotic therapy reduced the average diarrhea duration by 30 % (p = 10-4). Fever at inclusion was associated with a longer episode (HR = 0.61; 95 %CI: [0.41-0.89]; p = 0.012). No adverse event of medications was reported. CONCLUSION: A single dose of ciprofloxacin was effective and safe in treating uncomplicated diarrhea among service members in Africa.

Diabetes incidence in subjects with PREDIABetes from ReUNion Island: the PREDIABRUN observational cohort study protocol.

INTRODUCTION: Improved knowledge of prediabetic subjects' profile and their risk of developing type 2 diabetes mellitus (T2DM) would enhance secondary prevention. The primary objective is to describe factors associated with incident T2DM in subjects with pre-diabetes diagnosed in primary care. METHODS AND ANALYSIS: The study is based on Reunion Island, a French overseas region that experiences a particularly high disease burden of T2DM. This is an observational, non-randomised prospective cohort study conducted in primary care in which private general practitioner (GP) investigators recruit participants with pre-diabetes from their practices regardless of the initial motive for consultation. Pre-diabetes is defined by WHO criteria, that is, fasting plasma glucose between 1.10 g/L and 1.25 g/L and/or plasma glucose 2 hours after ingestion of 75 g of glucose (2-hour post load plasma glucose) between 1.40 g/L and 1.99 g/L. The design is based on an annual follow-up by the GP (according to French National Health Authority recommendations) with collection of clinical and laboratory data and specific lifestyle questionnaires answered by telephone at three time points: inclusion, and at 2-year and 5-year follow-up visits. Follow-up clinical and laboratory data are collected by the investigating GP as part of the study, and study-specific laboratory collections (serum, DNA and urine) will be obtained 2 and 5 years after inclusion. The primary outcome is transition to T2DM. ETHICS AND DISSEMINATION: This protocol has been approved by the research ethics committee of Saint Etienne (CPP Saint Etienne reference number: 2019-03). Enrolment began in August 2019. Results will be disseminated in at least three papers published in peer-reviewed medical journals, one oral communication and a large-scale communication to the local population and healthcare policymakers. TRIAL REGISTRATION NUMBER: NCT04463160 and ID-RCB 2018-A03106-49.

Nelfinavir in HIV-HCV coinfected patients: a 24-month follow-up in a cohort of 82 patients.

This retrospective and longitudinal study evaluated the long-term hepatic tolerance of a nelfinavir (NFV)-antiretroviral combined regimen in 82 patients of the HCV-HIV Cohort of CISIH-Sud of Marseilles. Follow-up data (liver enzyme levels, CD4 cell count, HIV viral load, and metabolic parameters) of patients treated with NFV on inclusion or during the follow-up of the cohort were analyzed under treatment over 24 months. Comparisons were performed with X2 or Kruskal-Wallis tests. At baseline (n = 82), the median exposure to NFV was 4.1 months; 58 patients received NFV combined with NRTI and 24 with NNRTI. The median CD4 cell count was 337/mm3 [interquartile range (IR): 216-480) and 39.7% had an undetectable HIV RNA level. Qualitative HCV PCR was positive in 91% of the patients and 19/51 patients with liver biopsy were F3-F4. Median alanine and aspartate aminotransferase (ALAT, ASAT), gamma-glutamyltransferase (GT), and alkaline phosphatase (ALP) were 46 UI/liter (IR: 36-76), 55 UI/liter (IR: 32-97), 97 UI/liter (IR: 50-194), and 88 UI/liter (IR: 72-104), respectively, with 76% of the patients with ALAT/ASAT grade <2. Median follow-up was 23 months (IR: 13.8-37). No significant difference was observed in the distribution of ALAT, ASAT, GT, and ALP as well as of ALAT/ASAT grades over the 24-month study period. Patients treated with NFV + NNRTI had significantly higher GT and ALP levels at baseline with no significant increase during follow-up. Cholesterol, triglyceride, and glycemia distributions remained stable over time. In conclusion, this study showed a good hepatic and metabolic tolerance of a long-term NFV-combined regimen in HIV-HCV coinfected patients.

Chikungunya infection: self-reported rheumatic morbidity and impaired quality of life persist 6 years later.

This study presents the 6-year follow-up of French gendarmes exposed to the chikungunya (CHIK) infection in 2006 on Reunion Island. The aim was to see to what extent the subjective health differences observed in 2008 (30 months after infection) between CHIK infected (CHIK+) and noninfected (CHIK-) gendarmes still persisted in 2012, and to investigate a possible return to a pre-CHIK health status for CHIK+ subjects. Gendarmes were contacted by mail in 2012 and asked to complete a self-questionnaire asking for morbidity, health care and medicines consumption since the last follow-up in 2008. Quality of life (QoL) after 6 years was evaluated using the SF-36 scale. In comparison with CHIK- subjects (n = 171), CHIK+ (n = 81) presented with higher rheumatic but also nonspecific morbidity such as headaches and fatigue associated with a large psychological impact, frequent depressive moods and social disabilities, leading to a significant impairment of the QoL and higher health care consumption. When restricted to CHIK+ subjects, comparing the data with that of 2008 showed persistent but decreasing self-reported rheumatic morbidity, and an increase over time of chronic discomfort (headache, fatigue) and depressive moods, resulting in no overall improvement in QoL. Despite possible cohort attrition bias, the comparability of CHIK+/CHIK- subjects allows the assumption of a long-term impact of CHIK infection with less chance of returning to a previous health status. Although these results may be specific to the 2006 virus strain, we recommend that public health strategies in the epidemic-prone countries include a response to the consequences of chronic post-CHIK disorders.

Severe hepatic cytolysis: incidence and risk factors in patients treated by antiretroviral combinations. Aquitaine Cohort, France, 1996-1998. Groupe dEpidémiologie Clinique de Sida en Aquitaine (GECSA).

OBJECTIVE: To study hepatic cytolysis in patients treated by highly active antiretroviral therapy (HAART) with protease inhibitor or with two nucleoside reverse transcriptase inhibitors (NRTIs). METHODS: We selected patients of the Aquitaine Cohort who initiated HAART or two NRTIs before 1 January 1998, had alanine amino-transferase (ALT) < or = 200 IU/I at baseline and at least one follow-up measure. Cox model was used to study the association between occurrence of severe hepatic cytolysis (ALT>200 IU/l) and age, gender, HIV transmission group, baseline CD4 and CD8 cell count, history of hepatic cytolysis, antiretroviral drug, baseline liver enzymes (WHO classification level 0: < or = 50 IU/l, level 1: 51 to 100, level 2: 101 to 200), hepatitis B and C co-infection. RESULTS: Sixty-four of 748 (8.5%) patients treated with HAART and 71 of 1249 (5.7%) treated with two NRTIs developed cytolysis. The probability of occurrence was 7.9% after 1 year [95% confidence interval (CI), 5.9-10.4] for patients treated with HAART and 4.8% (95% CI, 3.6-6.4) for patients treated with two NRTIs (log-rank test, P = 0.01). The median time to occurrence was 164 days for HAART-treated patients and 252 days for those treated with two NRTIs. In multivariate analysis, the history of cytolysis [hazard ratio (HR) = 2.3; 95% CI, 1.2-4.4], baseline value of ALT (HR = 2.4; 95% CI, 1.2-4.8 and HR = 3.3; 95% CI, 1.4-7.4 for levels 1 and 2, respectively), hepatitis B (HR = 3.0; 95% CI, 1.4-6.2) and C co-infections (HR = 3.2; 95% CI, 1.7-6.2) remained significantly associated with the occurrence of severe hepatic cytolysis among HAART-treated patients. History of cytolysis, hepatitis B and C were associated with cytolysis in patients treated with two NRTIs (HR = 14.8, 2.6 and 2.7, respectively). CONCLUSION: Hepatic cytolysis is more frequent among patients treated with HAART than with two NRTIs. Hepatitis B and C are the major risk factors after initiation of HAART or treatment with NRTIs. Co-infections with hepatitis B virus or hepatitis C virus may modify the management of HIV-infected patients treated by HAART.

Incidence of acquired immunodeficiency syndrome (AIDS)-related Kaposi's sarcoma in the Aquitaine Cohort, France, 1988-1996. Groupe d'Epidémiologie Clinique du SIDA en Aquitaine.

OBJECTIVE: To assess secular trends of the incidence of Kaposi's sarcoma (KS) between 1988 and 1996 in the Aquitaine Cohort of human immunodeficiency virus type 1 (HIV1)-infected subjects (southwestern France). METHODS: Adults of both sexes of all HIV-transmission categories were included. We distinguished between incident and prevalent KS and in case of multiple acquired immunodeficiency syndrome (AIDS) defining illnesses between initial or subsequent KS. Only incident KS were considered for annual incidence rate calculation. RESULTS: Overall, 21.2% (356/1678) of homosexuals and 1.9% (58/3030) of the other patients were diagnosed with KS over time. Although there was a sharp decrease in 1996 for initial KS, the annual incidence rate of KS was stable over time in the overall cohort as well as in homosexuals (4.3% per year on the average for KS as an initial AIDS-defining illness and 2.1% per year for subsequent KS in homosexuals). The median CD4+ cell count at the time of diagnosis of KS was 56 per mm3 (78 for initial KS, 14 for subsequent KS), with no significant variation over time. CONCLUSION: In the Aquitaine Cohort, the annual incidence of KS has remained stable between 1988 and 1995 with a recent decline in 1996, only for initial KS, while case management of HIV-infected subjects changed drastically.

Incidence of primary opportunistic infections in two human immunodeficiency virus-infected French clinical cohorts.

BACKGROUND: Clinical guidelines for the prevention of opportunistic infections in human immunodeficiency virus (HIV)-infected individuals have been developed on the basis of natural history data collected in the USA. The objective of this study was to estimate the incidence of primary opportunistic infections in HIV-infected individuals in geographically distinct cohorts in France. METHODS: We conducted our study on 2664 HIV-infected patients from the Tourcoing AIDS Reference Centre and the hospital-based information system of the Groupe d'Epidémiologie Clinique du SIDA en Aquitaine enrolled from January 1987 to September 1995 and followed through December 1995. We estimated: (1) CD4-adjusted incidence rates of seven primary opportunistic infections in the absence of prophylaxis for that specific infection or any antiretroviral drugs other than zidovudine; and (2) CD4 lymphocyte count decline. RESULTS: The highest incidence rates for all opportunistic infections studied occurred in patients with CD4 counts < 200/microl. With CD4 counts < 50/microl, the most common opportunistic infections were toxoplasmic encephalitis (12.6 per 100 person-years) and Pneumocystis carinii pneumonia (11.4 per 100 person-years). Mycobacterium tuberculosis was the least common opportunistic infection (< 5.0/100 person-years). Even with CD4 counts > 300/microl, cases of Pneumocystis carinii pneumonia and toxoplasmic encephalitis were reported. The mean CD4 lymphocyte decline per month was 4.6 cells/microl. There was a significant association between HIV risk behaviour and the incidence of cytomegalovirus infection, between calendar year and the incidence of Pneumocystis carinii pneumonia, toxoplasmic encephalitis and Candida esophagitis, and between geographical area and the incidence of Pneumocystis carinii pneumonia and cytomegalovirus infection. CONCLUSIONS: Geographical differences exist in the incidence of HIV-related opportunistic infections. These results can be used to define local priorities for prophylaxis of opportunistic infections.

Benfluorex as a therapeutic option for insulin resistance in HIV lipodystrophy syndrome.

BACKGROUND: HIV lipodystrophy syndrome, characterized by a significant excess of visceral adiposity and a reduced subcutaneous fat mass in association with insulin resistance and dyslipidemia, still affects the majority of antiretroviral-treated HIV-infected patients. The therapeutic management of this syndrome has not yet been well established. Benfluorex is known to decrease insulin resistance with no side effects on lactate levels in HIV-negative patients. METHOD: We conducted an open-label study of benfluorex (150 mg, 2-3 times a day) that was prescribed for 60 HIV-infected patients who were diagnosed with glucose metabolism abnormalities by oral glucose tolerance test (OGTT); 47 of these patients had visceral fat accumulation measured by computed tomography (VAT). Median follow-up was 12 months (interquartile range [IQR] = 6-12 months). The great majority of patients (90%) were treated with at least triple therapy (in 70% the therapy included at least one PI), with a nonsignificant change over the study period. RESULTS: Added to antiretroviral therapy, benfluorex improved OGTT in 47/60 cases, including total normalization in 34/60 without lactate concentration modification. A trend toward a decrease in VAT distribution was observed (p =.06). No significant difference was observed in subcutaneous fat distribution, although an increase in subcutaneous thigh adipose tissue was observed in 17/47 (36.2%) cases and 6 patients (12.7%) presented both subcutaneous fat increase and VAT decrease.

Efficacy and tolerance of HCV treatment in HIV-HCV coinfected patients: the potential interaction of PI treatment.

PURPOSE: To evaluate tolerance and efficacy of an open-label interferon-ribavirin treatment and their determinants in 62 HCV-HIV coinfected patients in routine followup. METHOD: Patients received at least 6 and up to 12 months of combination interferon alpha-2b (peg or not) plus ribavirin. Determinants of therapeutic success were estimated by a multivariate logistic regression. RESULTS: Five patients stopped the study, 4 were lost to follow-up, and 53 participated in the entire therapeutic protocol. Among these 53, the end-of-treatment results showed complete clearance of HCV-RNA in 17 (32%). A sustained virologic response (SVR) after 6 or 9 months was observed in 9 (17%) patients, 3 relapsed, and data were not available for 5. Genotype 3a (odds ratio [OR] = 14.4; confidence interval [CI] = 1.84-110.3) favored SVR and treatment with protease inhibitor (PI) therapeutic resistance (OR = 14.4; CI = 1.01-200); as well, a higher fibrosis score tended to increase resistance (p =.11). Adverse events were reported by 24/53 patients (45.3%). CONCLUSION: HCV therapy associating interferon and ribavirin in HCV-HIV coinfected patients is well accepted even if tolerance is moderate. Treatment permitted SVR in at least 17% of the cases. This is likely when patients initiate treatment at the early fibrosis stage and are infected with genotype 3a. The potential interaction with PI therapy should be explored.

[Human immunodeficiency virus infection and AIDS in Aquitaine. 10 years' experience of a hospital information system, 1985-1995. Le Groupe d'Epidemiologie Clinique du SIDA en Aquitaine (GECSA)]. / L'infection par le virus de l'immunodéficience humaine et le SIDA en Aquitaine. Dix ans d'expérience d'un système d'information hospitalier, 1985-1995. Le Groupe d'Epidémiologie Clinique du SIDA en Aquitaine (GECSA).

OBJECTIVES: To describe the hospital-based information system developed by the GESCA (Clinical epidemiology group on AIDS in Aquitaine) and evaluate the main results obtained over the last 10 years, specifically data related to epidemiological monitoring, medico-economic analysis and clinical research in HIV infection. METHODS: Inclusion criteria were HIV-1 seropositivity confirmed by Western blot, age over 13 years, consultation or hospitalization is one of the 18 participating units. Follow-up data were collected at each hospital visit. RESULTS: On December 31, 1995, 4268 subjects had been registered in the GECSA system, including 25.7% women. Contamination categories were homosexuals 33.4%, intravenous drug abusers 32.8%, heterosexuals 13.9%. A clear drop in the number of new cases in intravenous drug abusers and an increase in the number with heterosexual transmission was observed after 1988. The medico-economic analysis showed that patients followed in the system were in a more advanced stage at registry since 1992. More than half of the subjects are now taking antiretroviral therapy and prophylaxis for opportunistic infections. Clinically, after adjusting data for the major known prognosis factors, the risk of progression to AIDS is higher in homosexuals and intravenous drug abusers than in heterosexuals. Transmission route does not however have any significant effect on survival after development of AIDS. There is no significant difference in outcome of HIV infection between men and women and pregnancy is not associated with poorer outcome. CONCLUSION: This regional registry provides valuable data for epidemiological monitoring, medico-economic analysis, hospital management and clinical research.

[Decreasing incidence of HIV infection in intravenous drug users in Aquitaine: 1985-1993]. / Décroissance de l'incidence de l'infection a VIH chez les toxicomanes intraveineux en Aquitaine: 1985-1993.

OBJECTIVE: To describe trends in the incidence of HIV infection diagnosis since 1985 in different at-risk groups in Aquitaine, south-west France. METHODS: We analyzed the data from two regional surveillance systems on HIV infection: the hospital-based system of the Groupe d'Epidémiologie Clinique du SIDA en Aquitaine (GECSA) and the laboratory-based system of the Observatoire Régional de la Santé d'Aquitaine (ORSA). The number of intravenous drug users (IVDUs) in Aquitaine was estimated from two sources: a study performed by the GECSA in IVDUs treatment centers (1990-1991 data) and the annual surveys undertaken by the Ministry of Health (SESI) in all the health facilities and social services of the region from 1988 to 1993. RESULTS: As of December 31st, 1993, 3478 HIV-infected patients had been reported to the GECSA, including 3031 Aquitaine residents. Among those, 1096 were IVDUs without any other at-risk behavior (36.2%), the largest transmission category. The annual incidence of HIV diagnoses has steadily and considerably changed over time for IVDUs: 175 new cases diagnosed in 1985, 218 in 1987, then falling to 38 in 1993. During the same period, there was no marked decrease for homosexuals and the incidence of diagnoses of heterosexually-acquired HIV infections increased. Between 1989 and 1993, the number of HIV tests performed by the laboratories involved in the ORSA surveillance system increased steadily from 80,310 to 176,250 per year, when the number of first positive tests in Aquitaine residents decreased from 744 to 319 per year (-57%). This decrease was greater for IVDUs (-80%) than for the other groups. The survey performed by the GECSA estimated that 990 IVDUs were followed annually by treatment centres in Aquitaine in 1990-1991. This figure was stable and in agreement with the results of the SESI surveys performed throughout the period. CONCLUSION: The incidence of new diagnoses of HIV infection has been decreasing in Aquitaine since 1988, particularly among IVDUs, suggesting that the epidemic is slowing down. It is essential to maintain these systems of epidemiological surveillance as a component of the HIV prevention program.

[Causes of hospitalization and death in the MANIF 2000 cohort, composed of HIV-infected patients contaminated by intravenous drug use, 1995-1999]. / Motifs d'hospitalisation et mortalité dans la cohorte MANIF 2000, composée de sujets infectés par le VIH et contaminés par usage de drogues en intraveineux, 1995-1999.

OBJECTIVE: To describe the causes of hospitalization and mortality among the MANIF 2000 cohort study, composed of HIV-infected patients contaminated through intravenous drug use. METHOD: Data collection with a standardized questionnaires with clinical, biological, therapeutic and psycho-sociologic data on inclusion and every six months. Dates and causes of hospitalization and death were collected retrospectively by nurses from the hospital clinical records. Comparison were made using the chi(2) or Mann-Whitney tests with 5% significance threshold. RESULTS: The MANIF 2000 cohort study included, between October 1995 and June 1998, 467 patients with a median age of 33 years, 30% of whom were women. On inclusion, 42.2% were still injecting drugs (half of them were on substitution therapy), 10.5% had stopped injections of drugs, 32% exhibited more than 500CD4/mm(3) and 55.7% had not taken antiretroviral treatment. As of December 31(st) 1999, 21 patients had died, i.e. a mortality rate of 19% persons/years of follow-up (10-fold greater than that expected in the general population). Less than 5 deaths were due to HIV (n=4). Suicides and liver disease each represented the same number of deaths. Out of the 335 patients not having missed more than one follow-up throughout the 24 month period, 120 had been hospitalized at least once (n=223 hospitalizations), i.e., a hospitalization rate of 2.8 per 100 persons/month of follow-up. A quarter of hospitalizations were due to benign infections or stage B or C pathologies according to the 1993 classification of HIV infections and one hospitalization out of 5 was due to psychiatric problems (in majority depressive syndromes). Other predominant causes were worsening of general state of health, trauma, problems related to alcohol consumption, drugs abuse or hepatic decompensation. CONCLUSION: HIV-infected patients contaminated by intravenous drug use represent a particular population that cumulates many risk factors and which requires careful monitoring of co-morbidities such as hepatic and psychiatric diseases in order to avoid premature death.

Determinants and evolution of squamous intraepithelial lesions in HIV-infected women, 1991-2004.

This study presents a case-control nested analysis of cervical squamous intraepithelial lesions (SIL) in a cohort of 423 HIV-infected women with registered Pap smears between 1991 and 2004. Data on Pap smear results, CDC HIV classification, CD4 cell count and antiretroviral therapy were prospectively collected. Pap smears were classified using the Bethesda classification. Women had a median of three Pap smears registered in the database. The first Pap smear was registered

Anthropometric indices as predictors of survival in AIDS adults. Aquitaine Cohort, France, 1985-1997. Groupe d'Epidémiologie Clinique du Sida en Aquitaine (GECSA).

The aim of the study was to assess the performance of weight related nutritional markers [reported involuntary weight loss (WL) greater than 10%, measured WL and body mass index (BMI)] in predicting survival at AIDS stage. The three anthropometric indices were used as time dependent variables in Cox models to predict survival at AIDS stage. The studied sample included 630 HIV1-infected individuals of a prospective cohort of those 421 died (median survival at AIDS stage: 19.9 months). After adjustment for usual prognostic factors of survival, the reported WL greater than 10% was a pejorative predictor of survival (hazard ratio (HR) 2.4; 95% confidence interval (CI): 1.9-3.0). For measured WL < 5%, between 5 and 10% and > or = 10% of baseline weight compared with no WL, HR were respectively, 1.9 (CI: 1.4-2.6), 3.3 (CI: 2.4-4.4) and 6.7 (CI: 5.2-8.6). The HR of death were 2.2 (CI: 1.6-3.0) for BMI between 16 and 18.4 kg/m2 and 4.4 (CI: 3.1-6.3) for BMI < 16 compared to normal BMI (> or = 18.5). Even a limited WL measured at a given point in time during follow up increases the risk of death at the AIDS stage. Simple cross-sectional measures of BMI have a good predictive value of survival.