the University of Washington, School of Medicine and Department of Epidemiology
    Public Health—Seattle and King County, HIV/AIDS Epidemiology Program
    Virginia Mason Medical Center, Division of Hematology/Oncology, Seattle, WA

    ABSTRACT

    METHODS: We used patient data abstracted from the Adult/Adolescent Spectrum of HIV-Related Diseases study of Public Health—Seattle and King County.

    RESULTS: Patients diagnosed with KS in the HAART era (n = 40) were significantly more likely (P < .05) than pre–HAART-era KS patients (n = 366) to be diagnosed with alcohol problems (43% v 18%), noninjection drug use (45% v 18%), injection drug use (25% v 10%), psychosis (25% v 13%), and hypertension (13% v 2%). Although median CD4+ count and HIV-1 viral load at the time of KS diagnosis were not significantly different between the two groups, significantly fewer (P < .01) HAART-era KS patients developed opportunistic illnesses (OIs) during their follow-up. The risk of dying among KS patients diagnosed in the HAART era is significantly lower (P < .01) than for KS patients diagnosed in the pre-HAART era (hazard ratio, 0.24).

    CONCLUSION: Although HAART-era KS patients in King County were as likely to have a depleted CD4+ cell count and high HIV-1 viral loads at the time of KS diagnosis as pre-HAART KS patients, they survived longer and fewer of them were diagnosed with other OIs. They also had an increased prevalence of substance abuse and mental illness, contributing to a dynamic and changing KS clinical profile.

    INTRODUCTION

    On a microscopic level, KS lesions are rife with activated endothelial cells, spindle cells, infiltrating leukocytes, and angiogenic activity.7 Clinically, patients may present with multicentric mucocutaneous plaques and nodules. Less frequently, KS may involve the viscera. Treatment with highly active antiretroviral therapy (HAART) has led to reduced morbidity in patients infected with HIV and partial or complete regression of AIDS-associated KS in many patients.8,9 In the United States,10-12 including King County, Washington,13 the incidence of KS has declined dramatically since the introduction of HAART.

    The rapidly changing landscape of HIV medical care has resulted in a change in the natural history of KS. Nonetheless, KS remains the most common HIV-related neoplasm10,12 and one of the most common opportunistic illnesses.14 In this study we sought to compare the clinical characteristics of KS patients in the HAART era with pre–HAART-era KS patients and with patients with AIDS who have never been diagnosed with KS.

    METHODS

    On enrollment onto ASD, an HIV-infected patient's demographic information and mode of HIV exposure are recorded as well as any previous diagnoses of conditions in the AIDS Surveillance case definition.16 Abstractors also obtain other health information from an enrollee's medical records for the 12-month period preceding enrollment (baseline period). Medical records are then abstracted regularly at 6-month intervals (follow-up periods) after the baseline period.

    We defined HAART as a drug regimen including any combination of three antiretroviral drugs, with or without a protease inhibitor. Among study patients with HIV or AIDS and a CD4+ count ≤ 500 cells/mL or HIV RNA viral load ≥ 10,000 copies/μL, prescription of HAART in Seattle and King County increased from approximately 15% in 1996 to 60% in 1997.13 Diagnoses were therefore defined as being in the HAART era if they were made during or after 1997. Any diagnosis made between 1990 and 1996 was defined as occurring in the pre-HAART era.

    We analyzed data collected between January 29, 1990, and June 1, 2003. AIDS was defined by the 1993 Centers for Disease Control and Prevention (Atlanta, GA) AIDS case definition.16 For patients enrolled onto the study before 1993, the 1993 AIDS case definition was applied retrospectively. Patients without any AIDS-defining infection or malignancy may have been diagnosed with AIDS at the time of their first CD4+ lymphocyte cell count less than 200 cells/μL or CD4+ less than 14%. KS diagnoses were determined by histology or by examination of lesions by an experienced AIDS clinician, thus meeting the AIDS surveillance case definition.16 For each opportunistic infection or neoplasm, only the initial date of diagnosis was used in calculating incidence. For dates where only month and year were available, the midpoint of that month was used as the date. Likewise, for dates in which only the year was available, the midpoint of that year was used as the date. CD4+ cell count and HIV-1 RNA viral load closest to a date of diagnosis were determined by selecting the measurement before or on the date of a particular diagnosis. If a CD4+ cell count or viral load measurement was not recorded in the 6 months before the date of that diagnosis, the closest date of CD4+ count or viral load was selected up to 6 months after the date of that diagnosis.

    Other clinical conditions or findings were documented as present in a particular study patient if the medical records available to the ASD abstractor explicitly stated the particular diagnosis or if that diagnosis could reasonably be inferred from statements in the medical records.17 Alcohol use problems were therefore recorded if the medical record indicated alcohol abuse or a history of treatment for alcoholism, including attendance at Alcoholics Anonymous. For ASD abstractors to record depression as a diagnosis for a particular study patient, a clinician would have to state specifically that diagnosis in the medical record and it would have to be severe enough to warrant antidepressant therapy. A study patient self-reporting that he or she was depressed was not adequate to record a diagnosis of depression. Injection drug use was recorded as a diagnosis if any intravenous, intramuscular, or cutaneous injection of a nonprescribed fluid was indicated in the medical record. Noninjection drug use was assumed and recorded for any abuse or use of illicit or nonprescribed drugs that was detailed in the medical record and which was not specifically characterized as injected. Hypertension was only recorded if the medical record documented it as an active problem requiring treatment. Abstractors did not record a history of hypertension, a family history of hypertension, or hypertension that did not require treatment. Psychosis was recorded if diagnosis of a severe emotional disorder was explicitly indicated in the medical record and if that disorder was not attributable to any infectious or neoplastic cause. Weight loss greater than 10 lb or 10% of the study patient's baseline weight was recorded if specifically stated in the medical record by the clinician or if calculated by the abstractor. Abstractors recorded smoking status only at the baseline interval. To be recorded as a smoker for this analysis, a documented history of smoking at the baseline interval was required. To be recorded as never smoking also required specific documentation in the medical record. Diagnoses of other clinical conditions were coded by the abstractors according to the International Classification of Diseases, ninth revision (ICD9).18 For the purposes of our analysis, we defined lipid disease to include any diagnosis of atherosclerosis (ICD9 414.0, 440.0 through 440.9, 437.0), isolated hypercholesterolemia (ICD9 272.0), isolated hypertriglyceridemia (ICD9 272.2), or mixed hyperlipidemia (ICD9 272.4). A study participant diagnosed with atherosclerosis or one of these lipid abnormalities was assumed to maintain that diagnosis during all subsequent follow-up. We also defined thrombosis as infarction (ICD9 410.0 through 410.9, 573.4), embolism (ICD-9 444.0 through 444.9, 453.0 through 453.9, 415.1), or vascular occlusion (ICD-9 433.0 through 433.9, 434.0 through 434.9, 362.3, 452, 436, 362.84, 557.0, 593.81).

    For analysis, we used information gathered from the 12-month baseline period as well as any subsequent information available from follow-up data abstraction forms. Statistical analyses were done using Statistical Analysis Software, Version 8 (SAS Institute, Cary, NC). We used the {chi}2 test to determine significance of associations and P values less than .05 were defined as statistically significant. We also used the Student's t test to determine significance of continuous data. To ensure that differences uncovered in the HAART era relative to the pre-HAART era of selected comorbid conditions were not due to differences in length of observation time available, we analyzed data using incidence and incidence density (incidence/person-time). Survival analysis was done using a Kaplan-Meier proportional hazards regression analysis.

    RESULTS

    The proportion of AIDS non-KS diagnoses represented by people of color increased significantly in the HAART era (17% v 31% for African Americans and 9% v 18% for Latinos; P < .01). Although the proportion of KS diagnoses in the HAART era did not demonstrate a statistically significant racial or ethnic redistribution, a similar trend toward increasing incidence of KS diagnoses among African Americans and Latinos relative to whites was observed (8% v 18% for African Americans, 10% v 13% for Latinos, and 78% v 65% for whites). Related to this trend in race distribution is the significant increase (P = .04) in the proportion of KS diagnoses made in the HAART era among patients originating from a country other than the United States (8% pre-HAART v 18% in the HAART era).

    The median patient age at the time of KS diagnosis did not significantly change in the HAART era compared with the pre-HAART era (37 v 36 years). In both periods, the majority of KS patients were diagnosed between the ages of 30 and 39 years.

    In the HAART era, patients who were infected with HIV lived significantly longer before developing AIDS and/or KS (data not shown). Among patients who eventually developed KS, the mean time between diagnosis of HIV infection and diagnosis of AIDS increased significantly (P < .01) from 3 years in the pre-HAART era to 5 years in the HAART era. In addition, the mean time between diagnosis of HIV infection and diagnosis of KS increased from 4 years pre-HAART to 7 years in the HAART era (P < .01).

    Patients diagnosed with KS in the HAART era live longer than patients diagnosed with KS pre-HAART. The 1-year survival rate for pre-HAART KS patients after diagnosis was 59%, whereas 79% of HAART-era KS patients survived at least 1 year after diagnosis (P = .03). In addition, using a Kaplan-Meier proportional hazards regression analysis, we found that the risk of dying among patients diagnosed with KS in the HAART era is significantly (P < .01) lower than if diagnosed with KS pre-HAART (hazard ratio, 0.24; 95% CI, 0.12 to 0.48). The median survival time for pre-HAART KS patients was calculated to be 16 months (95% CI, 14 to 19 months). In comparison, with a follow-up time of up to 60 months, HAART-era KS patient survival had reached a plateau at 74% and the median survival time cannot yet be calculated (Fig 2).

    HAART-era KS patients were significantly less likely (P < .01) to develop other Centers for Disease Control–defined opportunistic illnesses (OIs) during their time of follow-up as compared with pre–HAART-era KS patients (84% v 60%; Table 2). Pneumocystis carinii pneumonia is still the most common OI for KS patients to develop in the HAART era, even though its incidence declined significantly (50% pre-HAART v 25% in the HAART era, P < .01). Mycobacterium avium complex incidence also declined significantly (P < .01) in the HAART era (30% pre-HAART to 10% in the HAART era) and decreased from the second most common OI of KS patients to the sixth most common OI. The spectrum of other OI diagnoses and their relative incidence changed only slightly.

    Despite a decreased incidence of various OIs among HAART-era KS patients, the median CD4+ count (40 v 53 cells/μL) and median HIV-1 viral load (160,000 v 161,000 copies/mL) at the time of KS diagnosis were not significantly different from KS patients of the pre-HAART era (Table 1).

    When selected comorbid conditions are compared, persons diagnosed with KS in the HAART era were significantly more likely (P < .05) than persons diagnosed with KS in the pre-HAART era to ever have been diagnosed with alcohol use problems (43% v 18%), noninjection drug use (45% v 18%), injection drug use (25% v 10%), psychosis (25% v 13%), and hypertension (13% v 2%; Fig 3). In both eras, depression among KS patients was common, with prevalence rates of 57% pre-HAART and 63% post-HAART, respectively.

    Compared with HAART-era AIDS patients without KS, those with KS were significantly more likely (P < .05) to use noninjection drugs (45% v 27%), to be diagnosed with psychosis (25% v 7%), and to experience a thrombotic event (10% v 2%) (Fig 4). Ninety-five percent of HAART-era KS patients received a prescription for HAART at some point in their course of follow-up compared with 78% of AIDS patients without KS (P = .01). Importantly, a large portion (58%) of HAART-era KS patients did not receive a prescription for HAART before their diagnosis of KS (data not shown).

    Smoking among AIDS patients with and without KS in both eras was common, with prevalence among the various groups ranging between 58% and 70%. In contrast, a diagnosis of lipid disease was rare in both eras, with prevalence of between 5% and 8% among the various groups.

    Statistical differences between the various groups with respect to comorbid conditions did not vary with analysis of data using incidence density (incidence/person-time; data not shown).

    DISCUSSION

    Our findings also suggest that patients in King County diagnosed with KS in the HAART era had an increased prevalence of mental illness and substance use relative to both pre–HAART-era KS patients and HAART-era non-KS AIDS patients. It is difficult to explain the higher prevalence of mental illness and substance use among contemporary KS patients. Fortunately, these conditions do not seem to represent an absolute barrier to accessing health care, given that 95% of HAART-era KS patients received a prescription for HAART at some point during their time of follow-up.

    We speculate that HAART-era KS patients, perhaps associated with their higher prevalence of mental illness, substance use, and racial minority status, might have delayed accessing HIV-directed health care until their disease progressed to a more outwardly obvious and symptomatically evident condition. This seemed to be true for the 58% of HAART-era KS patients who were not prescribed HAART at any time before their KS diagnosis. This delay in treatment put them at risk for developing OIs—even in this modern era of HIV care. Late diagnosis of HIV, including HIV diagnosed at the same time as KS or another opportunistic illness, however, likely also contributed to the lack of HIV care before a KS diagnosis.

    It is noteworthy that 42% of HAART-era KS patients received a prescription for HAART before their KS diagnosis but nonetheless developed KS. One reason this occurred may have been poor adherence to whatever antiretroviral regimen they were prescribed. Evaluating the impact of substance use and mental illness on patient adherence is complicated. Additional analyses are needed.

    Another factor that could help explain why some study patients in the HAART era developed KS despite receiving HAART lies in the possibility that they may have acquired multidrug-resistant HIV and are thus less likely to reap the full benefits associated with such therapy. Although HAART is helpful in restoring the immune system and dramatically reducing HIV-1 viral loads, it does not ensure that KS will not develop. KS may occur in rare patients with CD4+ T-lymphocyte counts more than 200 cells/μL and nondetectable HIV-1 viral loads.20,21

    Recently, Nasti et al22 compared clinical data on 211 individuals with AIDS-related KS from two well-described Italian cohorts of patients with HIV. Fifty-one patients were already taking HAART at the time of KS diagnosis; 160 patients were naive to HAART at KS diagnosis and initiated such therapy after diagnosis. HAART-naive patients had a median CD4+ count of 61 cells/μL, which is similar to the median CD4+ counts of 53 and 40 cells/μL found, respectively, among pre–HAART- and HAART-era KS patients in our study. In contrast, Italian patients who received HAART before their KS diagnosis demonstrated a significantly higher median CD4+ count of 129 cells/μL. Similarly, those who received HAART before their KS diagnosis were more likely to have suppressed HIV-1 viremia than their HAART-naive counterparts or pre–HAART- or HAART-era KS patients in our study. HAART-era KS patients in our study include both HAART-naive patients and patients who have been prescribed HAART before their KS diagnosis. The finding that their CD4+ cell count and HIV-1 viral load at the time of KS diagnosis more closely matches those of the HAART-naive patients in the Nasti et al22 study emphasizes our belief that there are delays in accessing or adhering to appropriate HIV-directed health care in King County.

    Nasti et al22 found that improved immune function and virologic status had a significant effect in the presentation of KS at the time of diagnosis. Patients who had received HAART before diagnosis of KS presented with a less aggressive and more localized disease than those who were HAART naive. Although they found no difference in 3-year survival rates between the HAART-naive patients and the groups of patients previously treated with HAART, their results suggest that improving patients' access and adherence to HIV-directed care before a diagnosis of KS can have a positive impact in reducing KS-associated morbidity.

    We have not yet addressed the long-term impact that HAART prescriptions have had on CD4+ cell counts and HIV-1 viral loads, which is important to consider in seeking to gain a fuller clinical profile of modern KS patients and how they respond to therapy. This represents a limitation to this study—the extent to which CD4+ cell counts and HIV-1 viral loads are recorded in ASD is highly variable from patient to patient, and the timing of these tests does not necessarily correlate closely with other significant events (eg, date of AIDS diagnosis, date of KS diagnosis, and so on). Other limitations to this study include those inherent to a medical record review project such as the ASD study (limitations of ASD have been documented previously10) and the use of pre-existing data from a project not designed explicitly to explore trends in KS. For example, the dosage of drugs and the reasons for prescription are not recorded. In addition, although ASD encompasses a large and diverse portion of the HIV-positive population in Seattle and King County, it does not represent the entire population of HIV-infected persons in the area. Compared with HIV/AIDS Surveillance reports in King County, ASD captures an estimated 31% of HIV and AIDS patients in the Seattle and King County area.23,24 Similarly, it is not known what percent of all HIV-associated KS patients ASD captures and if there are any biases introduced by limiting our sampling of KS patients to those followed by ASD.

    The high prevalence of smoking among HAART-era KS patients in King County is disconcerting. We have shown that HIV-infected patients in King County who developed KS lived almost twice as long before they were diagnosed with AIDS. Mortality among AIDS patients with OIs has also dramatically declined in King County during the HAART era.15 Thus, with KS patients living longer and with a large fraction of them smoking, tobacco-related health problems, such as emphysema, lung cancer, and cardiovascular disease, may become more prevalent.

    Although the incidence of KS has declined dramatically since the advent of HAART, KS remains an important disease among the AIDS population. In Seattle and King County, a large portion of HAART-era KS patients represent a marginalized sector of society whose problems with mental illness and substance abuse may interfere with their ability to access HIV-directed health care. It may also be that the higher prevalence of mental illness and substance use among HAART-era KS patients contributed to poor antiretroviral drug adherence. The increased prevalence of mental illness and drug use among HAART-era KS patients will likely continue to present a challenge in securing long-term quality care for this group. Consequently, even in the era of HAART, KS patients continue to be afflicted with other OIs. Often, it is the OI that leads the patients to seek HIV treatment. The decline in KS diagnosis, the significantly increased time between HIV infection and development of KS, and improved survival following a diagnosis of KS, nonetheless, are testaments to advancements in HIV care.

    Authors' Disclosures of Potential Conflicts of Interest

    NOTES

    Supported by the Medical Science Research Training Program at the University of Washington School of Medicine.

    Presented in part as a poster at the 45th Annual Meeting of the American Society of Hematology, San Diego, CA, December 6-9, 2003, and in part as a poster at the 8th International Conference on Malignancies in AIDS and Other Immunodeficiencies, Bethesda, MD, April 29-30, 2004. Presented orally at the 2004 Western Student Medical Research Forum.

    Authors' disclosures of potential conflicts of interest are found at the end of this article.

    REFERENCES

    1. Goedert JJ, Cote TR, Virgo P, et al: Spectrum of AIDS-associated malignant disorders. Lancet 351:1833-1839, 1998

    2. Dukers N, Rezza G: Human herpes virus and epidemiology: What we do and do not know. AIDS 17:1717-1730, 2003

    3. Gao SJ, Kingsley L, Hoover DR, et al: Seroconversion to antibodies against Kaposi's sarcoma-associated herpes virus-related latent nuclear antigens before the development of Kaposi's sarcoma. N Engl J Med 335:233-241, 1996

    4. Engels EA, Biggar RJ, Marshall VA, et al: Detection and quantitation of Kaposi's sarcoma associated herpes virus to predict AIDS-associated Kaposi's sarcoma. AIDS 17:1847-1851, 2003

    5. Kramer A, Biggar RJ, Hampl H, et al: Immunologic markers of progression to acquired immunodeficiency syndrome are time dependent and illness specific. Am J Epidemiol 136:71-80, 1992

    6. Renwick N, Halaby T, Weverling GJ, et al: Seroconversion for human herpes virus 8 during HIV infection is highly predictive of Kaposi's sarcoma. AIDS 12:2481-2488, 1998

    7. Offermann MK: Consideration of host-viral interactions in the pathogenesis of Kaposi's Sarcoma. J Acquir Immune Defic Syndr 21:S58-S65, 1999

    8. Lebbe C, Blum L, Pellet C, et al: Clinical and biological impact of antiretroviral therapy with protease inhibitors on HIV-related Kaposi's sarcoma. AIDS 12:F45-F49, 1998

    9. Portsmouth S, Stebbing J, Gill J, et al: A comparison of regimens based on non-nucleoside reverse transcriptase inhibitors or protease inhibitors in preventing Kaposi's sarcoma. AIDS 17:F17-F22, 2003

    10. Jones JL, Hanson DL, Dworkin MS, et al: Incidence and trends in Kaposi's Sarcoma in the era of effective antiretroviral therapy. J Acquir Immune Defic Syndr 24:270-274, 2000

    11. Jacobson LP, Yamashita TE, Detels R, et al: Impact of potent antiretroviral therapy on the incidence of Kaposi's Sarcoma and non-Hodgkin's lymphomas among HIV-1-infected individuals. J Acquir Immune Defic Syndr 21:S34-S41, 1999

    12. International Collaboration on HIV and Cancer: Highly active antiretroviral therapy and incidence of cancer in human immunodeficiency virus-infected adults. J Natl Cancer Institute 92:1823-1830, 2000

    13. Public Health—Seattle & King County: Biannual Adult Spectrum of HIV-Related Diseases (ASD) Report. September 2003, 27

    14. Kaplan JE, Hanson D, Dworkin MS, et al: Epidemiology of human immunodeficiency virus-associated opportunistic infections in the United States in the era of highly active antiretroviral therapy. Clin Infect Dis 30:S5-S14, 2000

    15. Farizo KM, Buehler JW, Chamberland ME, et al: Spectrum of disease in persons with human immunodeficiency virus infection in the United States. JAMA 267:1798-1805, 1992

    16. Centers for Disease Control and Prevention: 1993 Revised Classification System for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Rep 41:1-19, 1992

    17. Centers for Disease Control and Prevention: Adult/Adolescent Spectrum of HIV Disease Surveillance Initiative: Form Completion Instructions. Bethesda, MD, CDC, May 2000

    18. World Health Organization: International Classification of Diseases, 9th Revision. Manual of the International Classification of Diseases, Injuries and Causes of Death, Vol 1. Geneva, Switzerland, WHO, 1997

    19. Washington State/Seattle—King County: HIV/AIDS Epidemiology Report. First half 2003

    20. Ball SC: Kaposi sarcoma and its changing course in HIV infection. AIDS Reader 13:470-479, 2003

    21. Martin-Carbonero L, Barrios A, Saballs P, et al: Pegylated liposomal doxorubicin plus highly active antiretroviral therapy versus highly active antiretroviral therapy alone in HIV patients with Kaposi's sarcoma. AIDS 20:1737-1740, 2004

    22. Nasti G, Martellotta F, Berretta M, et al: Impact of highly active antiretroviral therapy on the presenting features and outcome of patients with acquired immunodeficiency syndrome-related Kaposi sarcoma. Cancer 98:2440-2446, 2003

    23. Public Health—Seattle & King County: Biannual Adult Spectrum of HIV-Related Diseases (ASD) Report. July 2002

    24. Washington State/Seattle—King County. HIV/AIDS Epidemiology Report. March 2003

查询更多Patients相关信息在本站>>

《临床肿瘤学医学期刊》2005年2月第23卷第2期