Mediastinal Mass, Cancer, or Infection, Atypical Paraesophageal Actinomycosis Infection, Clinical and Surgical Approach.

Actinomyces israelii (AI) is a Gram-positive, rod-shaped bacterium that lives commensally on and within humans as a typical colonizer within the gastrointestinal tract, including the mouth. As an opportunistic pathogen, infection often results from tissue injury or breach of the mucosal barrier (ie, during various dental or GI procedures, aspiration, or specific pathologies such as diverticulitis). Symptoms generally present slowly as a non-tender, indurated mass that evolves into multiple abscesses, fistulae, or draining sinus tracts without regard for anatomical barriers, including fascial planes or lymphatic drainage. However, it may also present as an acute suppurative infection with pain and rapid progression to abscess formation.

Effects of reported beta-lactam allergies on pneumonia outcomes in lung transplant recipients.

BACKGROUND: The effects of reported beta-lactam allergies on clinical outcomes have been understudied in lung transplant recipients. We evaluated reported beta-lactam allergies on clinical outcomes in this population. METHODS: A single-center retrospective cohort analysis was performed. One hundred and nine lung transplant recipients were identified and screened for a diagnosis of pneumonia. This cohort was divided into those with a reported beta-lactam allergy and those without a beta-lactam allergy. Antibiotic use was compared between groups. We also compared several clinical metrics, including rates of readmission, mortality, Clostridium difficile infection (CDI), allograft dysfunction, and isolation of carbapenem or fluoroquinolone non-susceptible organisms after treatment. RESULTS: Of the 109 lung transplant recipients, 18 (16.5%) were identified as having a reported beta-lactam allergy. Patients with a beta-lactam allergy label (BLAL) were found to have decreased utilization of beta-lactams (p < .001) and a trend toward increased use of carbapenems (p = .062) and aztreonam (p = .080). BLAL patients were found to have higher rates of CDI (p = .049) but were not found to have increased readmissions, mortality, allograft dysfunction, or isolation of carbapenem or fluoroquinolone non-susceptible organisms. Patients without a BLAL were found to have higher rates of acute kidney injury (AKI) (p = .035). CONCLUSIONS: Lung transplant recipients with BLAL are more likely to develop CDI, possibly due to increased use of carbapenems. We also found that patients without beta-lactam allergy were more likely to develop AKI. A multicenter study with a larger sample size might clarify the individual contributions of mutually confounding clinical parameters.

Infectious complications of bronchial stenosis in lung transplant recipients.

BACKGROUND: Bronchial stenosis is a known complication of lung transplantation, but there are limited data regarding whether transplant recipients with bronchial stenosis develop more infectious complications than those without bronchial stenosis. METHODS: We conducted a retrospective single-center observational cohort study between January 1, 2011 and September 29, 2016 of 35 lung transplant recipients diagnosed with bronchial stenosis and a random sample of 35 lung transplant recipients without bronchial stenosis. Data collected included donor/recipient demographic and anatomic information, respiratory cultures, episodes of respiratory infections diagnosed using CDC-NNIS criteria, hospitalizations, and 1-year all-cause mortality. Patients were followed up to 1 year after transplant. RESULTS: Bronchial stenosis occurred at a median of 54 days post-transplant (range 5-365 days). Bronchial stenosis patients spent more time in the hospital (87.4 vs 46.8 days, P = 0.011) and had more total hospitalizations (4.54 vs 2.37, P < 0.01) than their counterparts. The relative risk of pneumonia among cases vs controls was 4.0 (95% CI 2.2-7.3, P < 0.01); for purulent tracheobronchitis the relative risk was 3.1 (95% CI 1.6-6.1, P < 0.01). Patients with bronchial stenosis were significantly more likely to have respiratory cultures growing Staphylococcus aureus (RR 5.0; P = 0.001) and Pseudomonas aeruginosa (RR 2.1, P = 0.026). Mortality within the first year following transplant was equal in both the groups (14.3% vs 14.3%). CONCLUSIONS: There was no significant increase in 1-year mortality for lung transplant patients who developed bronchial stenosis. However, bronchial stenosis patients had significantly higher risks of pneumonia and tracheobronchitis, and spent more days in the hospital than those without bronchial stenosis.

Case-case-control study of patients with carbapenem-resistant and third-generation-cephalosporin-resistant Klebsiella pneumoniae bloodstream infections.

Strains of third-generation-cephalosporin-resistant Klebsiella pneumoniae (3GCRKP) and carbapenem-resistant K. pneumoniae (CRKP) are rapidly spreading. Evidence is needed to establish whether differences exist between patients at risk for 3GCRKP and those at risk for CRKP bloodstream infections (BSIs); thus, this retrospective case-case-control study was conducted to determine if the risk factors for these two infections differ. The inclusion criteria for cases were positive blood cultures for K. pneumoniae, first episode of BSI, age of ≥18 years, and susceptibility results indicating resistance to either third-generation cephalosporins (3GCRKP group) or carbapenems and cephalosporins (CRKP group). Controls were patients admitted for ≥72 h and were matched to cases by month/year and medical unit. Variables of interest were analyzed by univariate analysis, and those of significance were analyzed by logistic regression. In total, 111 patients with 3GCRKP BSIs and 43 patients with CRKP BSIs were matched to 154 controls. Multivariate analyses of 3GCRKP case and control groups demonstrated that a length of stay (LOS) of >40 days (odds ratio [OR], 17.7; 95% confidence interval [CI], 3.7 to 84.3), the use of antibiotics in the past 90 days (OR, 4.3; 95% CI, 1.5 to 11.9), and the presence of a central venous catheter (OR, 4.1; 95% CI, 1.3 to 13.4) were independent risk factors. Multivariate analyses of the CRKP case and control groups demonstrated that a LOS of >40 days (OR, 13.5; 95% CI, 2.9 to 62.8) and the use of antibiotics in the past 90 days (OR, 5.9; 95% CI, 1.3 to 26.5) were independent risk factors. Similar factors put patients at risk for these two types of K. pneumoniae BSIs.

An outbreak of norovirus infection in a bone marrow transplant unit.

BACKGROUND: Norovirus is a single-stranded RNA virus belonging to the Caliciviridae family. METHODS: Our observational cohort study aimed to describe a nosocomial outbreak of norovirus on a bone marrow transplant (BMT) unit. RESULTS: Six of 8 BMT patients with increased liquid stools tested positive for norovirus: 4 had new onset diarrhea; 2 had acute exacerbations of chronic diarrhea caused by graft versus host disease. Eight non-BMT inpatients had norovirus infection, but 7 of these were community acquired; cumulative incidence rates in BMT and non-BMT units were 26% and 0.16%, respectively. In BMT patients, diarrhea (increased or new onset) lasted 6 to 33 days-durations shorter than those reported in sporadic BMT cases. All patients had private rooms and bathrooms. Five of 6 patients were on the BMT unit during their presumed incubation periods. Three were in adjacent rooms. Three nurses and 1 physician had symptoms compatible with norovirus infection, and all 4 worked while ill. The outbreak ended coincident with implementation of stricter infection control practices. CONCLUSION: Norovirus appeared to spread in a BMT unit more avidly than it did among general medical patients. Explanations include prolonged diarrhea and viral excretion, long hospital stays of infected patients, rarity of empiric contact isolation for diarrhea, routine handling of liquid stool, and a closed community of health care workers.

Longitudinal epidemiology of multidrug-resistant (MDR) Acinetobacter species in a tertiary care hospital.

BACKGROUND: Acinetobacter species are well-known causes of health care-associated infections. The longitudinal epidemiology of this species in the hospital setting is poorly understood. A sudden, persistent increase in multidrug-resistant (MDR) A baumannii infections occurred beginning in June 2006 at Temple University Hospital in Philadelphia. An analysis was done to describe the longitudinal molecular epidemiology of MDR A baumannii in a tertiary care hospital. METHODS: This was an epidemiologic investigation using repetitive extragenic palindromic-PCR (rep-PCR) of patients with a positive culture for MDR A baumannii admitted to the hospital between February 2006 and January 2010. MDR A baumannii were defined as susceptible only to colistin and/or tigecycline. RESULTS: The incidence rate of MDR A baumannii rose from 0.36 cases per 1,000 patient-days (pre-epidemic) to 0.86 cases per 1,000 patient-days, due mainly to an increase in the surgical intensive care unit. Enhanced infection control measures were implemented, but waves of MDR A baumannii continued to be documented through routine surveillance. Of 32 strains collected in 2006-2007, a single predominant clone and 2 minor clones accounted for almost all of the cases of MDR A baumannii studied. Of 24 strains collected in 2008-2009, another clone, different from those studied in the earlier period, predominated, and was accompanied by 3 minor variants. CONCLUSION: Following an outbreak in the surgical intensive care unit, MDR A baumannii persisted in our institution for a 3-year period despite rigorous infection control measures. An unexpected strain replacement occurred during this period, with the original predominant strain disappearing completely and new minor clones displacing the original minor clones.

Prevalence and characteristics of patients with undiagnosed HIV infection in an urban emergency department.

The Centers for Disease Control and Prevention (CDC) recommends offering HIV testing to persons admitted to emergency departments (EDs). Whether by opt-in or opt-out, many EDs (including our own) have found a seroprevalence of 0.8-1.5% when rapid testing is offered. The true seropositivity rate is unknown. We performed a retrospective chart analysis upon all patients presenting to our ED over a 2-week period in the fall of 2007 who had serum drawn as a part of their emergency care. Demographics and clinical characteristics were linked via de-identified serum, which was sent for HIV testing. Nine hundred fifty nine patients had sera available for rapid HIV testing. One hundred twenty one (13%) samples were reactive via the OraQuick(®) test (OraSure Technologies, Bethlehem, PA), a point of care rapid antibody test. Due to concerns about the appropriateness of sera as substrate for the OraQuick(®) technology, reactive samples were retested via standard enzyme immunoassay (EIA)/Western blot. One hundred twelve analyzable samples were retested-38 were positive and 27 of these were from patients who reported a history of HIV infection. The rate of undiagnosed HIV infection was 1.2% (11/914 potentially analyzable samples). Of all patients with HIV in our ED, 29% of them were presumably unaware of their diagnosis. In conclusion, HIV seroprevalence in our urban ED is high, and a large fraction of the patients appears to be unaware of the infection.

Decreased length of stay and cumulative hospitalized days despite increased patient admissions and readmissions in an area of urban poverty.

BACKGROUND: Length of stay at US acute care hospitals has been steadily decreasing since 1960, and there is ongoing concern that increasing financial pressures on hospitals with high proportions of Medicaid patients may be causing unduly short lengths of stay. OBJECTIVE: To study temporal trends in hospital utilization on internal medicine services at Temple University Hospital, which has the highest percentage of Medicaid and uninsured patients in the state of Pennsylvania. DESIGN: Examination of temporal changes in hospital practice over three time periods spanning 13 years. MEASUREMENTS: Numbers of discharges, 1- and 12-month re-admission rates, and lengths of stay. US census data from 1990 to 2000 were examined for the eight major zip codes in which hospitalized patients live. MAIN RESULTS: The number of internal medicine admissions increased from 1991 (117/month) to 2004 (455/month); p < 0.0001. Mean length of stay for the index admission decreased from 8.7 to 4.9 days; p < 0.001. The percentage of patients readmitted within 12 months of the discharge date of the index admission increased from 42.3% to 49.5%; p = 0.045. Mean cumulative length of stay over 12 months, including readmissions, decreased significantly (15.8 to 12.5 days; p = 0.031). Compared to all US hospitals, our hospital had a greater increase in admissions and a greater decrease in length of stay. During this time period, in surrounding zip codes, there were decreases in total population and total number of persons living in poverty, but also multiple closures of area hospitals that served poor patients. CONCLUSION: During the 13-year study period, despite increased readmission rates, the overall number of hospitalized days per year on the internal medicine inpatient service decreased. As local hospitals serving this inner city low income area have closed, our hospital had atypically high increases in numbers of admissions and decreases in length of stay. This raises questions about current adequacy of hospital care in inner city areas of poverty.

Inflammatory cutaneous adverse effects of methylene blue dye injection for lymphatic mapping/sentinel lymphadenectomy.

BACKGROUND: Methylene blue (MB) dye has been used for lymphatic mapping/sentinel lymphadenectomy (LM/SL) in staging of melanoma and breast cancer. It has been noted to cause skin necrosis, but its more mild adverse effects from intraparenchymal breast injections are not well characterized. METHODS: Patients undergoing LM/SL for breast cancer and melanoma were reviewed, with attention devoted to skin manifestations. Patients undergoing mastectomies were excluded to rule out changes from flap devascularization. All breast patients were injected intraparenchymally. RESULTS: Ninety-five patients underwent MB injection during a nationwide shortage of lymphazurin; 78 for breast cancer and 17 for melanoma, with 51 patients undergoing breast conservation (BCT). There was no frank skin necrosis among any of the patients. Six (11.8%) BCT patients demonstrated inflammatory changes. Four patients developed findings indistinguishable from infectious cellulitis, with two developing skin telangiectasias prior to radiotherapy. Two patients had fat necrosis confirmed at the MB injection site away from the surgical site; one on imaging and one by biopsy. Most symptoms resolved after conservative management. CONCLUSIONS: MB dye may cause cutaneous changes more subtle than previously described. Physicians caring for patients having LM/SL using MB should be aware of these effects so that a proper differential diagnosis can be entertained postoperatively.

Virologic success in an urban HIV clinic: outcome at 12 months in patients who were HAART naïve.

BACKGROUND: Randomized controlled trials with highly active antiretroviral therapy (HAART) have demonstrated over 70% virologic success rates, although patients in an inner city HIV setting likely have lower virologic success. METHOD: We studied the outcome of all treatment-naive patients beginning HAART in our urban clinic in Philadelphia, Pennsylvania. The primary outcome was virologic success at 12 months for all patients who were initiated on HAART. Secondary outcomes included virologic success at 12 months for only those who remained in care and the determination of which demographics influenced virologic success. RESULTS: Between 2003 and 2005, 109 patients were initiated on HAART: 39% women, 79% African American, 17% Hispanic, median CD4+ count 120 cells/mm3, and HIV-1 RNA 4.9 log10 copies/mL. Twenty-two were lost to follow-up after HAART initiation. Of the 87 who remained in care, 41 maintained a HIV-1 RNA <400 copies/mL through 12 months on their initial HAART regimen. Emerging drug resistance was documented in 7 of 87 patients. NNRTI-based HAART was significantly associated with greater virologic failure due to emerging resistance compared to a PI-based regimen. CONCLUSION: Our retrospective study demonstrates the difficulties in administering successful HIV care to an urban population, and efforts to help patients overcome barriers to consistent medical care must be a priority.

Outbreak of necrotizing enterocolitis caused by norovirus in a neonatal intensive care unit.

OBJECTIVES: To investigate an outbreak of necrotizing enterocolitis (NEC) in a neonatal intensive care unit (NICU) and to identify the etiology, describe illness risk factors, and develop control measures. STUDY DESIGN: A retrospective case-control study was performed including newborns with NEC and newborns without NEC, examining demographic factors and exposures to medications, staff members, and procedures before illness. Stool samples from affected newborns were collected and tested for bacteria, parasites, and viruses. RESULTS: We confirmed a NEC outbreak in the NICU in January 1998 with 8 cases, including 2 deaths, clustered in time and space. Norovirus-like particles were identified in all available stools from cases; norovirus (NoV) was confirmed with reverse transcriptase polymerase chain reaction in 4 of 6 samples. NEC cases were younger, had lower Apgar scores, and received antibiotics longer than 25 control subjects. Three NICU health care personnel had more contact with cases than control subjects; 1 staff member recalled having gastroenteritis symptoms around the time of the outbreak. CONCLUSIONS: This report associates NoV with NEC. NoV appeared to precipitate NEC in predisposed infants. Spatial clustering and epidemiologic links between cases and a health care worker with gastroenteritis suggests that NoV should be investigated among the etiologies of NEC outbreaks and that interventions targeted to interruption of NoV transmission should be considered.

Delayed recognition of a pseudo-outbreak of Mycobacterium terrae.

Pseudo-outbreaks of mycobacteria are difficult to recognize because of long incubation periods for growth and species identification. We report our experience with one clinical microbiology laboratory that isolated a species of nontuberculous mycobacteria from 14 patient specimens. These specimens came from 12 patients at 2 hospitals over a 6-day period and included 6 different fluids or tissues. Because of the delay between mycobacterial specimen submission and growth in culture, the outbreak was not noted until more than a month later. Initial species determination by a reference laboratory indicated that these isolates were Mycobacterium fortuitum. One patient received treatment for presumed M fortuitum brain infection, and it was not effective in changing her clinical course. The isolates were sent to the Centers for Disease Control and Prevention (CDC) for identification and typing by pulsed-field gel electrophoresis. The CDC determined that the isolates were an identical strain of M terrae, thus confirming a pseudo-outbreak. Combining pseudo-outbreak isolates with those correctly identified initially as M terrae during the 6-day period in question, there were 22 samples from 20 patients with M terrae. Since the pseudo-outbreak, the number of cultures of M terrae in the clinical laboratory has returned to baseline levels without any specific intervention.

Increased Barrett's esophagus for the decade between 1991 and 2000 at a single university medical center.

Barrett's esophagus is being diagnosed increasingly in the United States. The aim of this study was to determine whether the increased diagnosis of Barrett's esophagus is due to endoscopic reporting and/or a truly increasing rate. This retrospective study reviewed 18,183 endoscopy reports at Temple University Hospital from January 1991 through December 2000. Annual rates of new cases of endoscopically suspected Barrett's esophagus were determined. Biopsy results were reviewed for the diagnosis of Barrett's esophagus (i.e., specialized intestinal metaplasia). Rates of Barrett's esophagus increased from 3.22 to 8.28 per 100 endoscopies (257%; P < 0.01) on endoscopy and from 0.67 to 2.76 per 100 endoscopies (412%; P < 0.01) on histology from 1991 to 2000. Twenty-four and seven-tenths percent (252/1020) of patients suspected at endoscopy to have Barrett's esophagus were confirmed by histology. This study demonstrates an increasing rate of new cases of suspected Barrett's esophagus on endoscopy and confirmed Barrett's esophagus on histology over the last decade. The endoscopic impression of Barrett's esophagus was about four times higher than the confirmed diagnosis of Barrett's esophagus (intestinal metaplasia) on histology.

Potential clindamycin resistance in clindamycin-susceptible, erythromycin-resistant Staphylococcus aureus: report of a clinical failure.

The erm gene product confers clindamycin resistance on Staphylococcus aureus. We report a clindamycin clinical failure where resistance developed on therapy in a D-test-positive strain. D tests of 91 clindamycin-susceptible, erythromycin-resistant S. aureus isolates showed that 68% of methicillin-susceptible and 12.3% of methicillin-resistant S. aureus strains were D-test positive.

Elective spinal surgery in asymptomatic HIV-seropositive persons: perioperative complications and outcomes.

STUDY DESIGN: This is a retrospective study consisting of medical records review and evaluation of pertinent radiographs. OBJECTIVES: In this study, complications and outcomes are reviewed in a series of HIV-positive patients undergoing spinal surgery. Surgery was performed for conditions that were unrelated to HIV infection. METHODS: From the period 1996 to 2000, we identified 10 seropositive patients who underwent spinal procedures for conditions that did not arise as a complication of HIV infection. A retrospective review was done, which included inpatient and out patient records. RESULTS: A total of 6 women and 5 men who were HIV-positive underwent spinal surgery during the time period reviewed. Surgery was performed for a variety of conditions, including lumbar disc herniaition, degenerative disc disease, cervical disc herniation, spinal fractures, and Arnold-Chiari malformation. The mean CD4 count was 279 cells/mm3 before surgery. All patients had at least 6 months of conservative management before surgery, except in the case of those who had spinal fractures. Two patients sustained postoperative complications: 1 patient developed a superficial wound infection, which was successfully treated with antibiotics, and 1 patient experienced prolonged fever (7 days) after surgery, which resolved without a clear diagnosis. All patients were improved after surgery (mean follow-up, 29.3 months) CONCLUSIONS: Many physicians have held a nihilistic approach, when it comes to the treatment of HIV-positive persons. However, our results in this small series of patients suggest that spinal surgery may be appropriate and can be performed with acceptable outcomes in selected patients.

Achalasia presenting after operative and nonoperative trauma.

Achalasia has been described following fundoplication and is attributed to vagal nerve damage during surgery. Similarly, other traumatic events to the distal esophagus may be linked to the development of achalasia. Operative and nonoperative trauma as a possible factor in the development of achalasia was studied. A retrospective analysis of patients with achalasia (n = 64) at our institution was performed. Collected data included age, gender, symptoms, and history of operative and nonoperative traumatic events. Comparisons were made to a group of patients with similar symptoms but normal esophageal manometry (n = 73). Achalasia was diagnosed by manometry in 125 patients over a 6-year period. All patients with complete medical records (n = 64) were studied. A history of operative or nonoperative trauma to the upper gastrointestinal tract prior to the development of symptomatic achalasia was present in 16 of 64 (25%). Significantly fewer patients (9.5%) with symptoms of dysphagia, but normal manometry and upper endoscopy, had precedent trauma to the upper gastrointestinal tract (P < 0.05). All cases of nonoperative trauma occurred in motor vehicle accidents. Cases of operative trauma included coronary artery bypass surgery (n = 4), bariatric surgery (n = 2), fundoplication (n = 3), heart/lung transplantation (n = 1), and others (n = 5). Patients with proven achalasia and a history of trauma were more likely to have chest pain (RR, 4.5; P = 0.012) but less likely to have regurgitation (RR, 0.51; P = 0.01) or nausea/vomiting (RR, 0.0; P = 0.27) than those without a history of antecedent trauma. In this series, significantly more patients with achalasia had a history of preceding trauma than did patients with similar symptoms and normal esophageal manometry. Following trauma, patients may be at increased risk for developing achalasia, possibly from neuropathic dysfunction due to vagal nerve damage. Patients with posttraumatic achalasia may have symptoms which differ from those of other achalasia patients.

An outbreak of mediastinitis among heart transplant recipients apparently related to a change in the united network for organ sharing guidelines.

OBJECTIVE: To describe an outbreak of mediastinitis in heart transplant recipients. DESIGN: Retrospective and contemporaneous cohort SETTING: Urban tertiary-care university hospital with a large cardiac transplantation program. PATIENTS: Heart transplant recipients. INTERVENTIONS: Modifications of donor harvest technique; procedures aimed at decreasing skin and mucosal bacterial colonization; strict aseptic technique in the intensive care unit; and aggressive policing of established infection control practices. RESULTS: In April 1999, mediastinitis rates among heart transplant recipients increased abruptly from a baseline of 6 cases per 100 procedures to sequential quarterly rates of 22, 31, and 50 cases per 100 procedures, whereas infection rates in other cardiac operations were unchanged. Bacteria causing these infections were multidrug-resistant "nosocomial" organisms. The epidemic occurred 2 months after a change in the United Network for Organ Sharing organ allocation algorithm. This change resulted in an increase in the duration of preoperative hospitalization from a median of 52 to 79 days (P = .008) and may have promoted prolonged hospitalization of patients with high illness severity. Aggressive multidisciplinary interventions were temporally associated with a return to preoperative mediastinitis rates without changing length of hospitalization prior to transplantation. CONCLUSIONS: Changes in organ allocation for transplant that prolong waiting time in the hospital and alter illness acuity may lead to increased rates of postoperative infection. Measures to limit bacterial colonization may be a helpful countervailing strategy.

Continuation of postpartum antiretroviral therapy in a cohort of women infected with human immunodeficiency virus.

A common dilemma for HIV-positive pregnant women is the issue of continuation or cessation of antiretroviral therapy (ART) postpartum. Current guidelines for ART during pregnancy offer no specific recommendations for postpartum ART care. The objective of this study was to ascertain characteristics that would predict cessation or continuation of ART postpartum. In this study, prenatal and medical clinic records were reviewed retrospectively for a cohort of 29 HIV-infected pregnant women who were seen in the Temple University High Risk obstetrics practice from 1997 to 1998. All women took ART during pregnancy, except for one who received i.v. AZT and nevirapine during labor. HIV-specific medical care was provided concurrently during the time of the woman's obstetrics visit by a nurse practitioner and a clinical nurse specialist in consultation with the physician. Factors that were included for review included race, age, use of ART at the time of pregnancy diagnosis, type of ART during pregnancy, CD4 count, HIV-1 ribonucleic acid polymerase chain reaction (RNA PCR) levels, current substance use, disclosure of HIV status to current partner, years of HIV infection, prior HIV infected child, and whether this was a first pregnancy. The two groups of women were divided between those who discontinued ART postpartum and those who continued ART. The data were analyzed with the Kruskal-Wallis test for two groups, or calculations of risk ratios with Fisher's exact test. Study results indicated that 15 out of 29 women (51%) continued ART postpartum. The significant factors for continuation included Latina ethnicity (risk ratio = 0.24, confidence interval = 0.06-0.87), CD4 < 200 mm3 (p = .04), and a greater number of drugs in the antiretroviral regimen 3 versus 2 (p = .05). This study showed that postpartum continuation of ART was associated with identified Latina ethnicity, lower CD4 counts, and a greater number of drugs in the pregnancy regimen. Further study is recommended to understand the clinical impact of intermittent ART, the strategies for postpartum therapy adherence, and clinical follow-up.