Jeffrey T. Kirchner, DO

Current Centers for Disease Control and Prevention (CDC) estimates are that of the approximately 1 million persons in the United States infected with HIV, 250,000 to 300,000 are unaware of their serostatus. Therefore, they are not getting HIV-specific care or prevention counseling. It is this group, consequently, that significantly contributes to the progressing number (>40,000) of new infections seen annually in the United States. 

Physicians Ignoring Plea for Aggressive Testing

Despite the call to clinicians for the past several years for more aggressive testing, this has not been taking place. And despite the CDC's recommendation since 2001 for routine testing of all persons living in areas with a seroprevalence of at least 1%, testing continues to be underutilized.

The CDC data show that 41% of HIV-infected patients develop AIDS within 1 year of diagnosis.

In addition, we continually see a substantial number of persons diagnosed late in the course of their disease, when they present with an AIDS-defining illness or with very low CD4

Early identification of HIV infection provides physicians the opportunity to reduce transmission by effecting changes in risk behavior. Treatment with highly active antiretroviral therapy (HAART) decreases morbidity, mortality, and?in some settings?infectivity. The public health reasons for aggressive testing and early identification of infected persons are quite compelling. But it is apparent that many physicians outside of the HIV treatment world, including my own colleagues in family medicine, seem to ignore or perhaps are unaware of these data. Commonly cited "barriers" to lack of testing include time, expense, or lack of perceived HIV risk on the part of the patients or their physicians. 

The very recent publication of 2 studies in the February 10, 2005, issue of the 

 expands the call for routine HIV screening in the United States. In the first study, Dr Sanders and colleagues used a Markov model to evaluate the cost-effectiveness of a voluntary HIV-screening program in relation to issues of quality of life, survival, and the likelihood of sexual transmission.

The HIV transmission component of this study was based on viral load, knowledge of HIV status, and efficacy of counseling. 

Given a 1% prevalence of unidentified HIV infection, the authors determined that regular screening increased life expectancy by 5.48 days at a one-time cost of $194 per screened patient. Identifying patients earlier in the course of their disease resulted in a survival advantage of 1.5 years for the average HIV-infected patient. A one-time screening program would reduce the annual transmission rate by more than 20% compared with the absence of screening. 

The cost of one-time screening would be $41,736 per quality-adjusted life-year. If the community prevalence was more than 0.05%, the cost of screening would still be less than $50,000 per adjusted lifeyear, which is the commonly cited cost threshold for screening for other diseases, such as breast cancer, diabetes, or hypertension. The authors cite a variety of demographic, natural history, testing, and treatment variables that support their outcomes. 

In the second study, Dr Paltiel and colleagues used a computer simulation model of HIV screening and treatment that compared routine voluntary counseling and testing with current testing practices in 3 target populations.

These included "high-risk" (3% prevalence of undiagnosed infection)," CDC threshold" (1% prevalence, 0.12% incidence), and "US general" (0.1% prevalence, 0.01% incidence). Primary outcomes included quality-adjusted survival, costs, and cost-effectiveness. 

Similar to the first study, the authors determined that one-time screening for HIV antibodies was associated with earlier diagnosis (mean CD4

cell count of 210/?L vs 154/?L) and an improvement in average survival. Also in keeping with the findings of Dr Sanders and colleagues, the incremental cost-effectiveness was $36,000 per quality-adjusted life-year gained. Dr Paltiel and colleagues concluded that in all but the lowest risk population, routine, voluntary screening for HIV once every 3 to 5 years is justified on clinical and cost-effective grounds. 

Of particular relevance to primary care physicians, Dr Paltiel and colleagues point out that HIV infection meets all the US Preventive Services Task Force criteria for targeted screening: 

It has a long preclinical, asymptomatic phase.  

It is a condition for which there is effective treatment. 

So we now have strong evidence to support what many of us have been calling for during the past several years?markedly increased testing for HIV infection. I hope that the findings from these 2 studies will be accepted by policymakers and professional societies, such as the American College of Physicians or the American Academy of Family Physicians, to encourage and support clinicians in performing more widespread HIV screening as part of routine medical care. Of course, with screening must come a linkage to HIV treatment and prevention services. 

As Dr Sam Bozzette notes in an accompanying editorial, screening programs ideally should be based in health care settings to minimize the stigma that has always surrounded HIV testing.

Moreover, office-based testing can take advantage of the fact that about 80% of adults in the United States see a health care provider at least once annually. Amodel of sorts is already in place, showing that the majority of women receiving prenatal care (100% in our own program in Lancaster) undergo routine HIV testing as standard of care. 

As with this population, my expectation is that the acceptance rate for routine testing would increase over time among patients visiting their physicians for an employment physical examination or a visit for hypertension, cholesterol, or diabetes management, or even for an acute problem. 

A perceived downside to routine testing is that more patients will be identified who will need HIV therapy, thereby also increasing health care costs. This comes at a time when at least 10 states currently have waiting lists for their federal AIDS Drug Assistance Programs, and there has been little or no expansion of funding for these programs in the past few years under the Ryan White CARE Act. 

However, as we expand the use of high-tech interventions at very low thresholds, such as implantable cardiac defibrillators, treating persons with HIV disease at a cost of roughly $10,000 a year has repeatedly been shown as a cost-effective intervention. And this pales in comparison with treating major opportunistic infections or AIDS-related malignancies, which result in lengthy hospitalizations and carry a much greater risk of death. Arather remarkable presentation at the recent 12th Conference on Retroviruses and Opportunistic Infections in Boston determined that HIV care has saved nearly 2 million lives in the United States since 1989.

If we continue our failing to provide widespread HIV testing and subsequent identification of those infected, the numbers of newly infected patients in the United States will continue to expand, following the current patterns of the global AIDS epidemic. For a disease that has no cure and no effective vaccine in sight, this is a frightening prospect for our health care system to handle in the next decade and beyond. 

My plea to physicians: Talk to and encourage your patients to be tested for HIV at least once, regardless of whether you work in New York City, Kansas City, or Sioux City. Do not completely depend on the perception that your patient is not from a high-risk group or has never engaged in high-risk behaviors. As Dr Bozzette aptly notes, failure to pursue a policy of more widespread HIV testing is a disservice to our patients who are infected but unaware, their prospective contacts, and the public health of ourcountry.

Revised guidelines for HIV testing, counseling, and referral. 

Neal JJ. Frequency and predictors of late HIV diagnosis in the United States 1994-1999. Presented at the 9th Conference on Retroviruses and Opportunistic Infections, February 2002; Seattle, Wash [Abstract]. 

Sanders GD, Bayoumi AM, Sundaram V, et al. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. 

Paltiel AD, Weinstein MW, Kimmel AD, et al. Expanded screening for HIV in the United States?an analysis of cost-effectiveness. 

Bozzette SA. Routine screening for HIV infection?timely and cost-effective. 

Walensky RP, et al. Two million years of life saved. The survival benefit of AIDS therapy in the United States. Presented at the 12th Conference on Retroviruses and Opportunistic Infections, February 22-25, 2005; Boston, Mass [Abstract # 143LB]. 

Articles

Is It Time to Begin Routine Screening for HIV Infection?

Medical Director, Comprehensive Clinic for HIV, Lancaster General Hospital, Lancaster, Pa

      DENVER, Colo&#8212;Many believe that the Conference on       Retroviruses and Oppor?tunistic Infections has become the most       important scientific meeting on the topic of HIV/AIDS. More than 3900       clinicians/researchers attended this year&#8217;s meeting, 48% of whom were       international attendees. The conference included 6 major plenary lectures,       7 symposia panels, 9 oral poster discussions, and >900 abstracts on       HIV/AIDS. Several papers have important clinical implications for primary       care physicians who treat both HIV-infected as well as noninfected       patients.   

      Protease Inhibitors May Increase Susceptibility to MI   

      This update was reported (Friis-Moller N. Abstract       144) from an ongoing study known as D:A:D (Data Collection on Adverse       Events of Anti-HIV Drugs) that includes 23,437 patients in 11 cohorts from       the United States, Europe, and Australia. All patients (median age, 39       years; 24% women) are taking at least 3 antiretroviral (ARV) drugs,       including a protease inhibitor (PI) and/or a non-nucleoside reverse       transcriptase inhibitor (NNRTI). Data from this study, which started in       1999, have been presented at previous meetings and were published in the New       England Journal of Medicine (2003;349: 1993-2003).       The primary outcome of this study is       myocardial infarction (MI) risk with duration of ARV therapy.    

      Findings presented this year involved patients who       were followed through December 2005, with a median duration to last       follow-up of 6.3 years. A total of 79% of patients had been receiving a PI,       62% an NNRTI, and 93% a nucleoside reverse transcriptase inhibitor. The       model was adjusted for age, sex, previous cardiovascular disease, and       smoking. There were 345 MIs in 94,469 person-years, for an incidence of       about 4 MIs per 1000 person-years. For the entire D:A:D cohort, the risk of       MI actually fell by 50% between 1999 and 2004. However, when looking       specifically at patients who had been taking a PI, the relative risk was       16% for each year of therapy. This translates into a doubling of the risk       for MI over a 5-year period. Even after adjusting for lipid profiles, the       relative annual risk was 10% for patients taking PI therapy. Patients       taking an NNRTI were at no increased risk. The investigators concluded that       PIs appear to increase risk for MI, irrespective of traditional Framingham       risk factors, such as total cholesterol, triglyceride, and high-density       lipoprotein levels.    

      Until the reason for this association is clarified, it       is incumbent on clinicians to aggressively manage other cardiovascular risk       factors in their patients whose ARV regimen includes a PI. Physicians may       also want to consider non-PI regimens in patients at high risk for       cardiovascular disease.   

      Postexposure Prophylaxis Often Begins Too Late   

      This study (Kindrick A. Abstract 906) was conducted by       the National HIV/AIDS Clinicians&#8217; Consultation Center at the       University of California, San Francisco (UCSF). It has been known for       several years that rapid initiation of antiretroviral ARV therapy after       occupational exposure to HIV effectively prevents infection. The Centers       for Disease Control and Prevention (CDC) has also issued recommendations       for the use of ARV therapy to prevent nonoccupational HIV infection,       including sexual exposure (MMWR.  2005;54:1-20). The optimal time to       treatment for both occupational and nonoccupational exposure, based on       biologic evidence and animal data, is probably just a few hours, although       the guidelines state treatment can be given up to 72 hours after HIV       exposure.    

Physicians and patients are unaware of the existence of nonoccupational exposure guidelines and the specific protocols that should be followed. 

      The National Clinicians&#8217; Post-Exposure       Prophylaxis Hotline (PEPline) at the UCSF provides postexposure advice       based on current CDC/US Public Health Service guidelines. The investigators       reviewed all calls that involved questions about management following       sexual exposure received at the hotline from January 2004 through August       2005 and examined exposure characteristics, individuals exposed, sources,       and treatment recommendations. During the 18-month study, there were 12,895       calls to the PEPline, of which 880 (5%) involved sexual exposure; 168 calls       were from emergency department physicians, and 183 were from family       physicians. An almost equal number of men and women were exposed, although       76% of the sources were men.    

      The majority of the phone calls involved exposures       that would have qualified for postexposure prophylaxis. However, only 283       (32%) of the calls came within 24 hours, while 259 (29%) came between 24       and 72 hours, and 248 (28%) came after 72 hours. Another 10% occurred after       an unknown time period. Thus most calls were made well beyond the 24-hour       time frame during which optimal benefit can be expected. Specific reasons       for the delay could not be ascertained, and it is unknown whether the       patient or the care provider was responsible. Other studies are under way       to identify potential barriers to timely management. My personal opinion is       that both physicians and patients are unaware of the existence of       nonoccupational exposure guidelines and the specific protocols that should       be followed.   

 Circumcision and HIV Transmission: The Cutting Edge     

     In a plenary talk (Abstract 120), Thomas Quinn, MD, of       the National In?stitute of Allergy and Infectious Diseases/ National       Institutes of Health, and the Johns Hopkins School of Medicine in       Baltimore, noted that only about 20% to 25% of men worldwide are       circumcised, compared with >50% of men in the United States. There is       great cultural and geographic variability associated with this practice. Dr       Quinn cited data showing a decreased incidence of cervical cancer       and human papillomavirus in Jewish and Muslim women whose husbands were       circumcised. There is also a lower incidence of herpes simplex virus and       other sexually transmitted diseases (STDs) in circumcised men. However,       medical indications for circumcision are currently limited to phimosis and       balanitis. In addition, the complication rate from routine circumcision is       1% to 3%.    

      Population-based studies show that HIV transmission       and acquisition are associated with blood and genital viral load, stage of       disease, viral subtype, presence of STDs, and male circumcision. In many       Asian and African countries where circumcision rates are <20%, the       prevalence of HIV is considerably higher than in countries in the same       geographic regions where circumcision rates are >80%. Dr Quinn described       a systematic meta-analysis of 38 studies that found a &#381;50% reduction in HIV infection among       circumcised men compared with uncircumcised men. A 2-year study from Uganda       looked at male partners of HIV-positive women and found that 40 of 137       uncircumcised men became infected, compared with none of the 50 men who had       been circumcised. A randomized trial in South Africa (n = 1538)       demonstrated that circumcision resulted in a 61% reduction in HIV       transmission, even after controlling for behavioral factors. The results       were significant enough that the study was stopped early. Two other similar       trials are ongoing in Rakai, Uganda, and in Kenya. Preliminary data from       the first trial (Abstract 128) show a reduced incidence of HIV acquisition       among women whose husbands are circumcised.   

      Dr Quinn presented some biologic explanations for the       association between circumcision and HIV transmission. The penile foreskin       mucosa is rich in the HIV target cells that allow more efficient absorption       of the virus. During intercourse the skin is retracted, allowing exposure       to the mucosa of the unkeratinized foreskin. In addition, the internal       foreskin is more prone to abrasions and tears and subsequent infection by       HIV as well as by other STDs.   

      Dr Quinn concluded his talk by emphasizing that there       is epidemiologic, biologic, and clinical trial evidence that male       circumcision significantly decreases the risk of HIV acquisition and       transmission. He presented a model showing that approximately 16       circumcisions would need to be performed to prevent 1 case of HIV       infection. Based on an estimated $69 per procedure, the overall cost would       be about $1000 per case of HIV prevented. Public policy implications for       widespread circumcision in high-risk countries must consider cultural       norms, national and local laws, and religious traditions. Any       circumcision programs would also need to be integrated into other STD and       HIV prevention programs and not promote a false perception of security in       circumcised men that could undermine future prevention efforts.   

      Evidence Argues Against ARV Drug Holidays   

      There has been a growing interest over the past       several years in strategies that allow patients taking ARV therapy to       interrupt their treatment and take a so-called &#8220;drug holiday.&#8221;       Some early studies failed to prove any immunologic benefit with this       strategy. However, ongoing concerns over drug toxicity and costs have       prompted continued research. The results of 2 trials presented at this       meeting found that stopping ARV therapy may not be a wise strategy for       certain patients.   

      The SMART (Strategies for the Management of       Anti-Retroviral Therapy) study was an international randomized       controlled trial that began in January 2003 at 318 sites in 33 countries       (El-Sadr W. Abstract 106LB). The median age of patients was 46 years; 27%       of the participants were women.    

      Patients in the first arm of the trial who had a CD4+ count >350       cells/mm3 continued to receive ARV therapy, with a goal of maintaining an       undetectable viral load. Patients in the second arm, known as the drug       conservation group, were kept off medications until their CD4+ count fell below       250 cells/mm3. Therapy was again stopped if the CD4+ count went above 350       cells/mm3. The primary end point was death. The trial was designed to       enroll 6000 patients but was stopped in January 2006 after it was found       that patients in the discontinuation group had       a 2.5-fold increased risk of HIV disease progression or death, a 1.5-fold       increased risk of severe complications, and a 6.1-fold increased risk of       serious HIV disease progression. Of note, most of the deaths, including       those due to cardiovascular, hepatic, and renal disease, were not       specifically AIDS related.     

      A second study conducted in Cote d&#8217;Ivoire, West       Africa, ANRS 1269 Trivican, found similar results. This study included 840 adults who were randomized to continuous ARV therapy; 2 months of treatment followed by 4       months off treatment; or discontinuation of therapy if their CD4+ count was >350       and not restarting until the count was <250 cells/mm3. Similar to the       SMART trial, this study found that patients in the latter group who stopped       therapy with a CD4+ count <250 had a 2.6-fold increased risk of morbidity       and mortality, mainly due to serious bacterial infections. The third arm of       the study was stopped as well.      

      The take-home message from these trials is that       patients receiving stable therapy without drug side effects or       complications should be maintained on therapy. Of note, 3 other smaller       studies presented at the meeting found no significant mortality risk with       treatment interruptions, but their designs did not allow patients to stop       ARV therapy unless CD4+ counts were in the  range of 350 to 450 cells/mm3. I       suspect we will continue to see more treatment discontinuation or       intermittent therapy trials in the future, but with more specific and safer       immunologic criteria.   

      Normalizing HIV Testing in Health Care Settings   

      Timothy Mastro, MD, of the CDC, participated in a       session dedicated to public health strategies and harm reduction for HIV       prevention (Abstract 164). He described updated epidemiologic data showing       that between 1 million and 1.2 million persons in the United States are       living with HIV/AIDS, but about 25% are unaware of their status. When       individuals do not know they are infected, they are more likely to transmit       the virus to unsuspecting partners. These individuals are believed to be       responsible for about 60% of the 40,000       infections that occur in the United States each year. In addition, many       persons are tested late in the course of the disease when ARV therapy is       less effective. Finally, voluntary testing of all HIV-infected pregnant       women, which has been a CDC recommendation for >10 years, is still       inconsistently performed.   

      To increase knowledge of HIV status and help link       patients with prevention, medical care, and treatment services, the CDC       will be issuing new testing guidelines in late spring or early summer 2006.       Current CDC guidelines were issued in November 2001 and recommend screening       only in settings where seroprevalence is &#381;1%, targeted testing based on risk assessment, and yearly       testing for men who have sex with men. Dr Mastro pointed out several       reasons for the proposed changes in the guidelines, including that many       HIV-infected persons access health care but are not tested for HIV until       they are symptomatic (usually with an AIDS diagnosis), despite knowing that       very effective treatments are available. Awareness of HIV infection       substantially reduces high-risk sexual behavior. In?creasing knowledge       about HIV has lessened the need for pretest counseling. More clinicians       have had experience with HIV testing, including the use of rapid tests.       Finally, evidence about the ?preventive benefits of typical counseling       for persons who test negative has been inconclusive.   

      The new guidelines will recommend routine HIV testing       in a variety of health care settings. They will focus on the model of       &#8220;opt-out testing,&#8221; meaning that HIV testing will be performed       after notifying the patient, and consent would be inferred unless the       patient declines. Several studies, including a 2005 study from Texas, have       demonstrated that opt-out testing can increase the number of persons tested       by up to 59% and identify an additional 100 HIV-positive individuals       compared with an opt-in strategy (Texas Department of State Health       Services, 2005). There are similar data for prenatal settings, with one       study showing that 88% of women agreed to testing when an opt-out approach       was used compared with 35% with an opt-in approach (BMJ. 1999;318:      1660-1661). Dr Mastro also cited 2 recent studies published in the New       England Journal of Medicine (2005;352:570-585; 2005; 352:586-595) that       found routine HIV testing to be cost-effective even in settings where       prevalence was 0.1%.   

      The process for revising the testing guidelines       included an HIV prevention leadership summit in San Francisco in August       2005 and 2 meetings in Atlanta attended by both physicians and community       members. The preliminary recommendations include the following:   

      &#8226;	 Routine voluntary HIV screening in health care settings for all       persons aged 13 to 64 years, regardless of risk or prevalence   

      &#8226;	 Repeat screening, at least annually, for persons with known risk       factors   

      &#8226;	 Opt-out testing, with the opportunity to ask questions and the       option to decline; HIV consent included with general consent for care and       communication of test results   

      &#8226;	 Not requiring prevention counseling in conjunction with screening       in health care settings.     

      These guidelines are intended to apply to all health       care settings, including ?inpatient services, emergency departments,       urgent care clinics, STD clinics, tuberculosis clinics, community clinics,       and public health clinics. Dr Mastro noted that state and local regulations       vary and should be reviewed and revised as needed. In low-prevalence       settings, a &#8220;sunset&#8221; provision should be considered (ie, after       screening is initiated, if the prevalence of HIV in a community is found to       be <0.1%, ?continued routine screening could be ?discontinued).   

      Insurance reimbursement is not anticipated to be a       problem when HIV screening becomes more widespread. Adequate public funding       would also be expected to be available to support widespread testing.       Finally, assuring access to HIV care is critical if we are to identify the &#381;250,000 persons in the United       States with HIV whose status is unknown.    

      Dr Mastro emphasized that there will be further       discussion and debate before the guidelines are finalized. Clearly new       testing guidelines are overdue, and it is only through more widespread       routine testing that we can decrease the annual rate of 40,000 new       infections in the United States. 	  