Potent combination antiretroviral (ARV) therapies have significantly impacted clinical care and improved the prognosis of HIV-infected individuals. Their use, however, is not without complication. In addition to an array of short- and long-term ARV toxicities, immune reconstitution, or the reversal of HIV-related CD4 cell decline, may trigger an inflammatory reaction in some individuals soon after they begin anti-HIV therapy. Known as immune reconstitution syndrome (IRS), immune restoration disease (IRD), or immune reconstitution inflammatory syndrome (IRIS), this process can involve opportunistic infections (OI), malignancies, or inflammatory disorders. Collectively, these are believed to result from restored immune function in the setting of previously unrecognized antigenic stimuli. Since correctional physicians often have to start or re-start highly active antiretroviral therapy (HAART) for inmates/patients who did not receive or have access to such treatments before incarceration, IRS may occur in correctional settings with increasing frequency. This paper will discuss IRS with an aim to alert correctional practitioners to its implications. Readers may also be interested in several recently published review articles for a more in-depth and fully referenced discussion of IRS [1-3].

It has been suggested that IRS occurs in 20 - 30% of patients initiating HAART. However, the true incidence in an unselected population remains unclear because of the lack of prospective data and uniform IRS definition. Nonetheless, IRS appears relatively common. It seems that many events suspicious of IRS in routine clinical care, such as fevers or other mild complaints, may be self-limited and are never classified as disease specific IRS. The onset of IRS is not uniform, nor is it useful in distinguishing between disease specific IRS. The majority of IRS cases occur within two weeks to six months after HAART initiation, yet cases have occurred from as little as one day after HAART initiation, and up to several years later.

There is good evidence that lower nadir CD4 lymphocyte counts increase the risk of IRS. This is not unexpected in the face of the prevailing hypothesis that the presence of previously unrecognized pathogen or antigen triggers IRS. It should be noted that the relationship between lower nadir CD4 count and IRS is likely stronger for conditions that traditionally present at very low CD4 counts in AIDS patients. To date, no correlation between nadir CD4 count and the severity of IRS has been documented. In contrast to CD4 counts, pre-therapy HIV viral loads have not been predictive. It has been postulated that quicker decline of the viral load may be associated with an increased risk of IRS and there is evidence to support this. One should keep in mind that neither low baseline CD4 lymphocyte counts nor rapid lowering of HIV viremia is universally seen in patients with evidence of IRS, or that patients with either of these features uniformly experience IRS. Patient-specific demographic factors and mode of HIV transmission have not been associated with IRS, though this has not been well studied.

After initiation of HAART, IRS can arise in persons with or without a known history of prior OI. Although IRS presentation in these two settings may be clinically similar, there are unique questions that arise in the setting of recurrent disease. The first regards the appropriate timing of HAART in persons with pre-existing OI. This concept has been best studied in the setting of Mycobacterium tuberculosis (MTB) and HIV co-infection and will be discussed later. The second question regards the possibility of a recurrence due to the development of antimicrobial resistance. To date this has been largely a theoretical concern, however, in Pneumocystis jiroveci pneumonia (PCP) IRS some clinicians have opted to modify the existing treatment regimen without direct evidence for the presence of resistance. Based on other case reports, it does not appear that resistance was necessarily present. Finally, it must be remembered that a person with a pre-existing OI may have an IRS that is not related to that OI.

Because of the variability in presentation of particular IRS, the sections below will discuss aspects of individual IRS that are more common. The list is not exhaustive, and will focus on conditions that have classically been associated with advanced HIV disease. Other purported IRS have been reviewed recently [1-3]. Following the disease-specific IRS discussions, there will be a summary of what is known about the management and outcomes of IRS.

Mycobacterium avium complex (MAC). Atypical presentations of MAC infections during zidovudine monotherapy were the first evidence of IRS in HIV-infected patients prior to the HAART era. Since that time, MAC has been one of the most commonly described causes of IRS. Approximately 75% of MAC IRS represents new infection; the remainder arise in patients with a known history of MAC. The most widely described clinical presentation includes lymphadenopathy and fever; depending on the lymph nodes involved, focal painful syndromes or other symptoms are also common. Sites of lymphadenitis include cervical, mediastinum, abdomen, and the retroperitonium. Patients without lymphadenitis present with conditions including cutaneous abscess, granulomatous hepatitis, osteomyelitis, pyomyositis, hypercalcemia, and pulmonary infiltrates. Evidence that these cases have represented MAC IRS has commonly been obtained through biopsy and/or culture. In contrast to MAC infection with advanced AIDS, in MAC IRS, granulomatous inflammation is commonly recognized on histological examination. Tissue stains and cultures for acid-fast bacilli (AFB) are frequently positive, however, blood cultures for AFB are usually negative.

Mycobacterium tuberculosis (MTB). The majority of MTB IRS cases have arisen after HAART initiation in patients with tuberculosis (TB) disease. Because of the immunosuppressive nature of TB, at times, this clinical syndrome can be seen after the initiation of TB treatment alone in those with or without HIV infection. However, there is evidence that these reactions occur with a higher frequency in HIV-infected patients initiating HAART. Several trials have estimated MTB IRS to occur in 20-35% of HIV/TB co-infected patients and there is mounting evidence that earlier initiation of HAART is associated with a higher incidence of IRS in HIV/TB co-infected patients. In these complex cases, one needs to remember the potential for drug-drug interactions and overlapping drug toxicities as a cause of symptomatic complaints in patients on therapy for both TB and HIV. Whether or not to delay HAART must be individualized; there is little data on the overall risk of this strategy.

MTB IRS appears to be more common in patients with pre-existing disseminated (vs. isolated pulmonary) TB infection. It usually presents with fever and new or worsening lymphadenopathy, but many patients have multiple simultaneous complaints and exam findings. Other presentations may include worsening pulmonary symptoms (including acute respiratory distress syndrome), worsening symptoms from extra-pulmonary disease (abdominal pain, hepatosplenomegaly, headache, scrotal swelling, or cutaneous lesions), weight loss, diaphoresis, and hypercalcemia. Biopsies frequently show granulomatous inflammation but consistent findings regarding AFB stain, AFB culture, or MTB upon PCR have not been noted. Baseline tuberculin skin testing with purified protein derivative would not be expected to assist in predicting who will develop MTB IRS. The tuberculin skin test is frequently negative at baseline in patients who go on to develop MTB IRS, although many of these patients convert to a positive test during the period of immune reconstitution.

Cryptococcus neoformans. Cryptococcal IRS has most frequently been reported in the setting of pre-existing cryptococcal meningitis. More than 50% of patients with Cryptococcal IRS present with fever, headache, or both of these, and are found to have recurrent meningitis and/or increased intracranial pressure. Lymphadenopathy (mediastinal, supraclavicular, or cervical) is the second most common presentation followed by pneumonia, cutaneous disease, and hypercalcemia. Cultures of cerebrospinal fluid (CSF) or lymph node are usually negative; however, histology of excised lymph node typically shows

Pneumocystis Jiroveci Pneumonia (PCP). There are few published reports of PCP IRS, most of which have been seen in patients undergoing treatment for active PCP. It has been estimated to occur in 5 -19% of patients with PCP initiating HAART. In most reported cases, there appears to have been either inadequate length of steroid therapy or initiation of HAART prior to completion of PCP therapy. Clinical trials are ongoing to assess the appropriate timing of initiation of HAART in the setting of active OIs. Currently there are no official recommendations, although many clinicians wait until PCP therapy is complete prior to initiating HAART. Patients typically present with worsening symptoms of pneumonia (including respiratory failure) and fever. Atypical presentations have not been noted. Sputum or bronchoscopic specimen examination may be useful to rule out other pulmonary pathogens, but in most cases, identification of PCP does not help distinguish whether or not it is the cause of the pulmonary symptoms.

Cytomegalovirus (CMV). About two-thirds of CMV IRS reported in the literature occurs in the face of pre-existing CMV disease. Of all persons initiating HAART, it has been estimated that new CMV, as an IRS, occurs in about 5%. In contrast, up to 60% of patients with pre-existing CMV ocular disease have developed CMV IRS. Most patients present with visual changes and are diagnosed with ocular inflammatory lesions such as vitreitis. Extra-ocular disease has been reported, including colitis, pneumonitis, lymphadenitis, pancreatitis, parotitis, and febrile syndromes. In general, diagnosis has been achieved via direct ophthalmoscopic examination in patients with ocular disease. Of note, the site of primary CMV disease does not necessarily predict the site of recurrence. There is no data regarding the usefulness of serum PCR examination for CMV-DNA in diagnosing CMV IRS.

JC Virus (JCV). Because there are no specific antivirals available for the treatment of progressive multifocal leukoencephalopathy (PML) caused by JCV, HAART is the treatment of choice. In some individuals, however, PML commences or worsens after HAART. The symptoms of PML IRS are varied and typical of PML in untreated patients. These include paresis, dysphagia, visual changes, seizures, and ataxia. Most cases are diagnosed by magnetic resonance imaging (MRI) with or without biopsy or CSF-PCR analysis for JCV-DNA. In contrast to the pre-HAART era, when brain MRI rarely had inflammatory changes, in PML IRS, MRI commonly shows contrast enhancement consistent with inflammation. The inflammatory nature of these lesions has been verified in biopsy specimens showing peri-vascular inflammatory infiltrates, which were also very uncommon in the pre-HAART era. MRI, however, may not be helpful in predicting who will have a worsening course of PML after HAART initiation. There is evidence that most patients with a follow-up MRI have inflammatory changes while less than half of these patients have clinical worsening.

Hepatitis B Virus (HBV) & Hepatitis C Virus (HCV). It has been difficult to understand the contribution of Hepatitis IRS to abnormal liver function tests (LFTs) after HAART initiation in patients with HBV and/or HCV co-infections. Five to 10% of HAART responders have transaminase levels that reach five times the upper limit of normal and these changes have been more common in patients with chronic HBV or HCV infection. Although not definitive, liver biopsy changes in these patients have frequently been consistent with worsening viral hepatitis as opposed to drug toxicity. In general, LFT abnormalities return to baseline over several months but studies that have looked at this were not limited to patients with Hepatitis IRS as the cause of the LFT abnormality. 'Acute hepatitis', fulminant hepatic failure, and new onset porphyria cutanea tarda in an HCV-infected patient after HAART have been reported. Because of uncertainty in the cause of abnormal LFTs after HAART initiation, IRS and non-IRS causes need to be considered in this population.

Kaposi's sarcoma (KS). There has been a remarkable decrease in the incidence of KS in the post-HAART era. Like JCV, the virus associated with KS (human herpes virus 8, HHV-8), does not have effective antiviral options. Thus, HAART has become the treatment of choice and has been quite effective in many KS patients, although some still require chemo- or radiation-therapy. There have been reports of worsening of existing KS lesions, an increase in the number of KS lesions, or new onset KS in patients initiating HAART and these have frequently been associated with lymphadenopathy. One report showed a decrease in HHV-8 DNA by PCR analysis during the KS flare, which suggests that HHV-8-specific immune responses were reconstituted during this interval.

Management
There are no guidelines or evidence- based recommendations for IRS management. Several important issues need to be addressed. The first issue regards the continuation of HAART. Because IRS usually occur in the setting of effective therapy, it has generally been recommended to continue HAART except in extreme situations (CNS edema, tracheal or vascular compression, severe hepatitis, etc). Patients need to be counseled prior to HAART initiation and during IRS in order to make the best treatment decisions. In cases where HAART has been discontinued, IRS can recur during re-initiation of therapy, even in the face of specific anti-infective therapy. For these reasons, when possible, HAART should be continued.

A second issue regards the use of specific anti-infective agents (when available). In many cases, there is evidence of IRS resolution with continued HAART with or without anti-infective agents, making the decision of whether or not to initiate anti-infective therapy unclear. In selected circumstances, when a patient has partially completed treatment for a specific infection (e.g., TB infection), the prescribed course of therapy should be completed. The use of chemotherapy and radiation therapy may be required for KS IRS. It should also be noted that the occurrence of an OI after initiation of HAART does not always represent IRS. In patients with poor adherence and/or poor response to HAART, an OI may develop that is not an IRS and would need specific anti-microbial therapy when available. In IRS cases, individualized decisions need to be made based on the potential benefits and risks of specific anti-infective therapies. Prudence is warranted until more information is available.

A third and important issue is the use of anti-inflammatory agents. In many specific IRS there are reports of patients being treated with steroids or non-steroidal anti-inflammatory agents (NSAIDS). At this time it is unknown if either of these treatments should be employed in IRS. In several situations, the use of anti-inflammatory agents may be warranted. In severe or life threatening situations, particularly with pronounced inflammation, steroid use is likely appropriate. In patients with moderate to severe, non-life threatening events, consideration of either steroids or NSAIDS is reasonable, particularly if it can help patients continue their effective HAART. Finally, in situations where steroids have previously been proven beneficial, such as in moderate to severe PCP, use of these agents is likely warranted. Furthermore, depending on the IRS event, topical or intra-ocular anti-inflammatory agents may be useful. Many other situations may arise in which a decision regarding anti-inflammatory agents will need to be addressed and clinical judgment remains most important.

Outcomes
Although IRS can be severe or life threatening, in most instances, patients recover without major morbidity or mortality. However, because of the potential for severe complications, these events must be addressed quickly and appropriately. There have been reports of occasional severe ocular morbidity with CMV IRS, including blindness. At least one-third of patients with PML IRS have expired secondary to PML and most survivors have significant residual neurological deficits. Deaths have also been reported with MTB IRS and Hepatitis IRS. In all of these conditions, we await prospective clinical data to not only assist with finding effective and appropriate IRS management strategies, but also for IRS prevention strategies.

Conclusions
It is hopeful that some of these principles will be useful in helping clinicians understand, recognize, and manage IRS. There are many unanswered questions and research continues to help us address the many uncertainties. Recognition and management of IRS remains complicated and clinicians need to consider the many possible causes of symptoms that may or may not be related to HAART initiation. While it is comforting that IRS usually resolves with conservative management, the possibility of severe or life threatening reactions urges vigilance. Over the next several years there will likely be more and better information regarding IRS. Until then we will continue to rely on clinical acumen and the data synthesized from the published experience of many other clinicians.

Disclosures:
*Grant/Research Support: Bristol Myers Squibb

References:
1. Gardner EM, Connick E. Illness of Immune Reconstitution: Recognition and Management. Curr Infect Dis Rep 2004;6:483-493.
2. French MA, Price P, Stone SF. Immune restoration disease after antiretroviral therapy. AIDS 2004;18(12):1615-1627.
3. Shelburne SA, 3rd, Hamill RJ. The immune reconstitution inflammatory syndrome. AIDS Rev 2003;5:67-79.

Back to Top