Weekly Articles – 6/27/2021

Surial B, Mugglin C, Calmy A, Cavassini M, Günthard HF, Stöckle M, Bernasconi E, Schmid P, Tarr PE, Furrer H, Ledergerber B, Wandeler G, Rauch A; Swiss HIV Cohort Study. Weight and Metabolic Changes After Switching From Tenofovir Disoproxil Fumarate to Tenofovir Alafenamide in People Living With HIV : A Cohort Study. Ann Intern Med. 2021 Jun;174(6):758-767. doi: 10.7326/M20-4853. Epub 2021 Mar 16. PMID: 33721521.

I have written in the past about how integrase inhibitors are associated with increased weight gain and metabolic syndrome. This study uses data from the Swiss HIV cohort study to assess weight changes in those who were switched from TDF to TAF. Those who were on a TDF-based regimen for at least 6 months were included, as well as those who stayed on it and who were switched to TAF. Weight analyses were restricted to those who had 6 and 12 months of follow up, as well as those who had a viral load <50 copies/mL in the 12 months prior to the study. 4375 patients were included, of which 891 continued on TDF and 3484 (79.6%) were switched to TAF. Patients who switched to TAF were more likely to be men, have a lower median eGFR, and more likely to be on integrase-inhibitors as part of their ART. Crude weights diverged after initiation of TAF, with the TAF roup having an increase in 1.8kg over 18 months compared to 0.7kg in the TDF group (a between group difference, 1.1kg). 

After adjustment, these remained similar:

A higher proportion of patients in the TAF gropu became overweight or obese after 18 months (13.8%) compared to the TDF group (8.4%; difference 5.4, 95% CI 2.1 to 8.8). Switching to TAF was also associated with increase in total cholesterol, HDL, LDL, and triglyceride levels:

Have we been blaming the wrong drug all this time? Probably not, as patients who were treatment naive do demonstrate larger increases in weight when started on integrase inhibitors. Further, when looking at the integrase-inhibitor group, they had the biggest jump compared to all other variables:

Either way, consideration of cardiovascular risk reduction should be placed on anyone starting TAF + Integrase based regimens (such as BIC/TAF/FTC or DTG + FTC/TAF). 

Chammartin F, Lodi S, Logan R, Ryom L, Mocroft A, Kirk O, d’Arminio Monforte A, Reiss P, Phillips A, El-Sadr W, Hatleberg CI, Pradier C, Bonnet F, Law M, De Wit S, Sabin C, Lundgren JD, Bucher HC; D:A:D Study Group. Risk for Non-AIDS-Defining and AIDS-Defining Cancer of Early Versus Delayed Initiation of Antiretroviral Therapy : A Multinational Prospective Cohort Study. Ann Intern Med. 2021 Jun;174(6):768-776. doi: 10.7326/M20-5226. Epub 2021 Mar 16. PMID: 33721519.

We have known for some time that starting ART as early as possible (usually at diagnosis) confers the most benefit. Does this also translate towards cancers, specifically, AIDS-related cancers such as Kaposi’s or Non-hodgkin’s lymphoma? These next 2 articles seem to suggest so. This first study used data from 11 cohorts in 212 clinics in Australia, Europe, and the US. Here, the long-term risk for non-AIDS defining and AIDS-defining cancer after initiation of ART therapy, comparing 3 treatment strategies: 1) starting ART within 6 months regardless of any CD4 count, 2) starting ART within 6 months for counts <500 and 3) as above but for counts <350. Essentially, this was to emulate a pragmatic trial where 3 different treatment strategies could be evaluated so that those who had a CD4 count less than a threshold could be started on ART. A parametric g-formula was used to assess the 10-year risk for malignancy among all 3 strategies. 8318 patients meet includsion criteria. 489 patients developed a malignancy (231 had a non-AIDS defining cancer and 272 had an AIDS-defining cancer). The estimared risks for any AIDS-defining cancer at 10 years was 2.97% (95% CI 2.37 to 3.50) and for non-AIDS cancers was 2.50 (95% CI 2.37 to 3.38%) if ART was started immediately, and the risks increased with “delayed” ART initiation, with increasing absolute risk differences when ART was started at lower CD4 counts:

When looking at the subgroup of patients whose CD4 count was above 500 when starting ART, the overall risk of cancer was 2.15% for non-AIDS defining cancers and 1.38% for AIDS defining cancers, lower than the entire population.

Silverberg MJ, Leyden W, Hernández-Ramírez RU, Qin L, Lin H, Justice AC, Hessol NA, Achenbach CJ, D’Souza G, Engels EA, Althoff KN, Mayor AM, Sterling TR, Kitahata MM, Bosch RJ, Saag MS, Rabkin CS, Horberg MA, Gill MJ, Grover S, Mathews WC, Li J, Crane HM, Gange SJ, Lau B, Moore RD, Dubrow R, Neugebauer RS. Timing of Antiretroviral Therapy Initiation and Risk of Cancer Among Persons Living With Human Immunodeficiency Virus. Clin Infect Dis. 2021 Jun 1;72(11):1900-1909. doi: 10.1093/cid/ciaa1046. PMID: 32785640.

This study used data from the NA-ACCORD, a consortium of 22 single-site and multisite cohorts of HIV patients in the US and Canada. Here, observational data was used to evaluate early ART therapy (within 6 months) vs deferred ART therapy (after 7 months) in AIDS-free, cancer-free patients whose CD4 count was between 350-500 and their risk of developing either an AIDS-related or non-AIDS related malignancy. This study evaluated more of the “timing” of starting therapy once someone is linked to care, while the above study deals more with the hypothetical timing of infection to starting ART. Data up to 2014 was analyzed (before the guidelines recommended immediate initiation of ART). Data was analyzed in an ITT fashion, meaning once you were on ART you were assumed to be adherent to therapy. 21947 patients were evaluated, with 30% started therapy within 6 months, usually a median of 1.5 months after diagnosis. There were no significant differences between groups, although those who started ART earlier were enrolled later. The overall incidence of any cancer was lower in the early group compared to the deferred group:

With the rates of KS, NHL, and HL being lower in the early group, compared to lung and prostate. Adjusted HR demonstrated that the risk was lower for any cancer, but more specifically for the AIDS-defining cancer, for those who started on ART early:

Furthermore, early ART resulted in a lower cumulative incidence of cancer at all time points for any cancer, AIDS-related cancer, and any virus related cancer:

Both studies are quite large, though they do suffer from their observational nature, possible unmeasured biases, and the likelihood that we are learning more about HIV as the years go by. The mechanism behind the lower risk of malignancies (specifically AIDS-related malignancies) clearly relates to immune reconstitution, but the higher risk for non-AIDS related malignancies is baffling to me. Perhaps it is a similar explanation to the higher risk of MACE events in this population, where chronic HIV infection is a proinflammatory state, but I am not sure. 

E Moreno-García, P Puerta-Alcalde, G Gariup, M Fernández-Ruiz, L E López Cortés, G Cuervo, M Salavert, P Merino, M Machado, J Guinea, J García-Rodríguez, J Garnacho-Montero, C Cardozo, J Peman, M Montejo, J Fortún, B Almirante, C Castro, J Rodríguez-Baño, J M Aguado, J A Martínez, J Carratalà, A Soriano, C Garcia-Vidal, Project from GEMICOMED (SEIMC), Early Stepdown From Echinocandin to Fluconazole Treatment in Candidemia: A Post Hoc Analysis of Three Cohort Studies, Open Forum Infectious Diseases, Volume 8, Issue 6, June 2021, ofab250, https://doi.org/10.1093/ofid/ofab250

This was a retrospective post hoc study that included patients from 3 cohorts of patients with candidemia. Patients were divided into those who underwent early de-escalation of therapy to fluconazole (within 5 days) and those who did not. Those with fluconazole non-susceptible strains (except C. kruseii, which is always considered resistant, and C. glabrata, which is considered SDD to fluconazole) were excluded. 235 patients were included, of which 54 underwent early de-escalation at a median time of 3.5 days. Those who underwent early de-escalation were less likely to require ICU admission, had a known source of candidemia, and were less likely to be infected with C. glabrata. Mortality was significantly lower in the early de-escalation group at 30 days (11.1% vs 29.8%, p=0.006). Multivariate analysis found that COPD (OR 3.97, 95% CI 1,48-10.61), Pitt score >2 (OR 4.39, 95% CI 1.94-9.20), unknown source of candidemia (OR 2.59, 95% CI 1.14-5.86), and candidemia by C. albicans (OR 3.92, 95% CI 1.74-8.86) were associated with increased mortality. Early de-escalation to fluconazole (OR 0.50, 95% CI 0.16-1.53) was not associated with increased mortality. Similarly, multivariate analysis found that breakthrough candidemia and unknown source of candidemia were less likely to be associated with de-escalation. Pretty reassuring to know there is no impact on mortality if you de-escalate early, though given the great bioavailability of the azoles, I do not think this is quite surprising. Also, I still cannot get over the fact that C. krusei and C. glabrata were re-classified. I’m still calling them candida.

Elizabeth Ferzacca, Andrea Barbieri, Lydia Barakat, Maria C Olave, Dana Dunne, Lower Gastrointestinal Syphilis: Case Series and Literature Review, Open Forum Infectious Diseases, Volume 8, Issue 6, June 2021, ofab157, https://doi.org/10.1093/ofid/ofab157

This was an interesting case series of lower GI syphilis infection (i.e. that involving the rectom, colon, or anal canal). These were defined as a lower GI syndrome (anal mass or hematochezia) with serology or histopathology evidence of T. pallidum, and lower GI mucosal abnormality likely to be related to T. pallidum based on serology. 62 cases were evaluated, with most occurring in male patients, and 93% reporting having sexual contact with men. The most common symptoms included hematochezia, anal pain, abdominal pain, mucous discharge, diarrhea, constipation, and tenesmus: 

Findings included rectal mass, inguinal lymphadenopathy, and a rash:

Anorectal masses and ulcers were the most common endoscopic findings here:

Here are the listed histopathology findings, though I do not know what these really mean.

Nothing to comment here, though I am not surprised that syphilis can show up as rectal mass. The great mimicker strikes again. 

Julio Ramirez, Daniel H Deck, Paul B Eckburg, Marla Curran, Anita F Das, Courtney Kirsch, Amy Manley, Evan Tzanis, Paul C McGovern, Efficacy of Omadacycline Versus Moxifloxacin in the Treatment of Community-Acquired Bacterial Pneumonia by Disease Severity: Results From the OPTIC Study, Open Forum Infectious Diseases, Volume 8, Issue 6, June 2021, ofab135, https://doi.org/10.1093/ofid/ofab135

This one is an analysis of the OPTIC trial, which evaluated omadacycline vs moxifloxacin in those with CAP. Here, they performed additional analysis using 5 severity or mortality risk scores, as well as in those with COPD/asthma, bacteremia, and those with baseline radiographic findings. The original study was a phase 3 doublte-blind, multicenter, randomized trial evaluating omadacycline and moxifloxacin in patients with CABP POT risk class II, III, or IV and 3 pneumonia symptoms. They were randomized in a 1:1 fashion. Primary endpoint was clinical response measured at early clinical response (72-120 hours after first dose). Analysis at the posttreatment evaluation (5-10 days after the receipt of the last dodge of the study drug) was performed as a secondary endpoint. 774 patients were included in the ITT population. This image is a bit busy, but it essentially evaluates the end point when severity is stratified using different scoring systems. In other words, when looking at patients with PORT of IV at 72-120 hours after starting therapy (early clinical response), there was no difference in outcomes between antibiotics. Similarly, if you look at the post-treatment response using a SMART-COP score of >3, there was similarly no difference:

When looking at patients who had either baseline COPD/Asthma or baseline radiographic signs, both drugs performed equally well. 

The one exception was bacteremia, where omadacycline had lower success in both ECR (67% for omadacycline vs 89% for moxifloxacin) and at PTE (73% vs 83%). The authors do explain the difference in outcomes for bacteremia may stem from the small numbers evaluated (32 total patients). 

Choi SH, Dagher M, Ruffin F, Park LP, Sharma-Kuinkel BK, Souli M, Morse AM, Eichenberger EM, Hale L, Kohler C, Warren B, Hansen B, Medie FM, McIntyre LM, Fowler VG. Risk Factors for Recurrent Staphylococcus aureus Bacteremia. Clin Infect Dis. 2021 Jun 1;72(11):1891-1899. doi: 10.1093/cid/ciaa801. PMID: 32564065.

This is a single-center prospective cohort study evaluating patients with staphylococcus aureus bacteremia (SAB) which sought to identify characteristics associated with recurrent SAB. Recurrent bacteremia was defined as the second episode of SAB after resolution of the first episode, occurring at least 14 days from the last positive blood culture of the first episode. Spa genotyping was used for all SAB episodes, with recurrent SAB undergoing pulsed field gel electrophoresis. Those with indistinguishable pairs underwent whole genome sequencing. Multiple cytokines were also obtained for reasons. 759 patients with SAB were enrolled, of which, 72 (9.5%) had recurrent SAB. Of these, 69 patients were included, of which 55 had a single recurrence. Of the recurrent episodes, 41 had indistinguishable PFGE. The authors decided that if the time between episodes was >150 days, despite indistinguishable PFGE patterns, those would be considered re-infection. Overall, 30 were considered to have relapses while 39 had reinfections:

Patients with recurrent bacteremia were younger (56 vs 61), more frequently black (64% vs 30%), hemodialysis dependent (55 vs 16%), more likely to have a foreign body (83% vs 60%), have persistent bacteremia (39% vs 25%), and have MRSA (57% vs 44%). Multivariate regression analysis found that race and hemodialysis dependence were the risk factors associated with recurrent bacteremia:

Persistent bacteremia, MRSA, and metastatic abscess were more frequent in patients with relapsed SAB. While the authors test 25 cytokines, only one (RANTES) had significant differences between recurrent and single SAB episodes. I do not think this is clinically relevant. Further, I am not surprised by any of these results; if any one does any type of ID rotation, this should be fairly straight forward. Sticky staph is less likely to be cleared if there is a portal of entry or something foreign for it to stick.

Jasper AS, Musuuza JS, Tischendorf JS, Stevens VW, Gamage SD, Osman F, Safdar N. Are Fluoroquinolones or Macrolides Better for Treating Legionella Pneumonia? A Systematic Review and Meta-analysis. Clin Infect Dis. 2021 Jun 1;72(11):1979-1989. doi: 10.1093/cid/ciaa441. PMID: 32296816.

Macrolides vs Fluoroquinolones for Legionella! This was a meta-analysis that included 21 studies, and it evaluated macrolides and FQs for treatment for legionella pneumonia. Primary outcome was mortality. Most were observational studies (18/21) and only 3 were RCTs. A total of 3525 patients were included. 1636 were treated with FQ while 1889 were treated with macrolides. Mortality rates was 6.88% for FQ while for macrolides were 7.43%. Pooled OR for mortality for FQ vs macrolides was 0.94 (95% CI 0.71-1.25). There was no difference in mortality outcomes when looking at ICU-treated patients:

Whether or not you want to use FQs or macrolides will likely depend on the tolerability. While both increase the QTc interval, I would wager my bets on macrolides. Less adverse events, some immunomodulatory effects. 

Falcone M, Daikos GL, Tiseo G, Bassoulis D, Giordano C, Galfo V, Leonildi A, Tagliaferri E, Barnini S, Sani S, Farcomeni A, Ghiadoni L, Menichetti F. Efficacy of Ceftazidime-avibactam Plus Aztreonam in Patients With Bloodstream Infections Caused by Metallo-β-lactamase-Producing Enterobacterales. Clin Infect Dis. 2021 Jun 1;72(11):1871-1878. doi: 10.1093/cid/ciaa586. PMID: 32427286.

This was a study that evaluated CAZ-AVI + aztreonam compared to other antibiotic regimens for the treatment of bacteremia with metallo-beta-lactamase producing enterobacteriaceae. Aztreonam is not inactivated by MBLs but a lot of these bugs CAN produce other beta-lactamases that hydrolyzes ATM. Hence, the combination. This was a prospective study, with the primary outcome being 30-day all-cause mortality. 102 episodes of bacteremia were included, with 52 patients receiving the combination therapy while 50 received other active antibiotic therapy (usually a combination of colistin and another drug, or aztreonam + other drugs):

Patients who got the combination of CAZ-AVI + ATM were more likely to  be admitted to the ICU, less likely to have immunosuppressive therapy in the prior 30 days, and less likely to have drug-induced AKI. 30 day mortality was higher in the other antibiotic group (44%) compared to the CAZ-AVI + ATM group (19%). Mortality was significantly higher when looking at the colistin-based regimens (59.3%). Cox-regression analysis found that therapy with CAZ-AVI + ATM was protective of 30-day mortality:

This also held true when looking at propensity-score adjusted analysis (HR 0.31, 95% CI 0.15-0.66) as well as lower risk of clinical failure at day 14 (OR 0.36, 95% CI 0.18-0.7). While the duration of therapy was not statistically different between the two groups, I do not know if this played a role. As you can see, more patients in the colistin group died when compared to the non-colistin group, so whether drug toxicities also played a role is unclear. 

Son MBF, Murray N, Friedman K, Young CC, Newhams MM, Feldstein LR, Loftis LL, Tarquinio KM, Singh AR, Heidemann SM, Soma VL, Riggs BJ, Fitzgerald JC, Kong M, Doymaz S, Giuliano JS Jr, Keenaghan MA, Hume JR, Hobbs CV, Schuster JE, Clouser KN, Hall MW, Smith LS, Horwitz SM, Schwartz SP, Irby K, Bradford TT, Maddux AB, Babbitt CJ, Rowan CM, McLaughlin GE, Yager PH, Maamari M, Mack EH, Carroll CL, Montgomery VL, Halasa NB, Cvijanovich NZ, Coates BM, Rose CE, Newburger JW, Patel MM, Randolph AG; Overcoming COVID-19 Investigators. Multisystem Inflammatory Syndrome in Children – Initial Therapy and Outcomes. N Engl J Med. 2021 Jun 16. doi: 10.1056/NEJMoa2102605. Epub ahead of print. PMID: 34133855.

I do not write much about kids but here are 2 articles that deal with them. The first one describes the patterns of immunomodulatory meds used in those with multisystem inflammatory syndrome in children (MIS-C). Patients under 21 who were diagnosed with MIS-C were categorized based on whether they got IVIG only; IVIG and steroids; IVIG, steroids, and a biologic agent, or other treatments (any other combination of the other 3 therapies) at any point during their hospitalization. They also divided the patients based on what combination they got on day 0 (either IVIG plus steroids or IVIG alone). They reported multiple outcomes for those who got any immunomodulatory treatment during their hospitalization, but they used a pre-specified primary outcome of cardiovascular dysfunction (based  on a composite of LV dysfunction or shock) to assess initial therapy with either IVIG alone or IVIG plus steroids. 518 patients were evaluated. The cohort was actually quite sick with 74% requiring ICU admission, 55% having involvement of 5 or more organ systems, and 38% meeting diagnostic criteria for Kawasakii’s. The majority of patients got a combination of IVIG and steroids (47%). Most of the cohort that got all 3 types of medications during their admission were more likely to be critically ill. 

Using propensity score matching, 206 patients were matched in a 1:1 ratio for IVIG alone or IVIG + steroids on day 0. Initial treatment with IVIG + steroids was associated with a lower risk of the composite outcome of CV dysfunction on or after day 2 than IVIG alone (17% vs 31%, RR 0.56, 95% CI 0.34 to 0.94), with the risk of the components of the composite outcome also favoring the combination group. Similar findings were obtained using inverse-probability-weighted analysis, with initial therapy with IVIG + steroids being associated with a lower risk of composite outcome:

While they do note that those who got steroids + IVIG upfront were less likely to require further immunomodulatory therapy, a significant portion (72%) of the IVIG alone propensity-matched cohort required some sort of immunomodulatory therapy at some point in their admission. Whether this played a role or not is difficult to say. 

Wilke J, Ramchandar N, Cannavino C, Pong A, Tremoulet A, Padua LT, Harvey H, Foley J, Farnaes L, Coufal NG. Clinical application of cell-free next-generation sequencing for infectious diseases at a tertiary children’s hospital. BMC Infect Dis. 2021 Jun 11;21(1):552. doi: 10.1186/s12879-021-06292-4. PMID: 34112116; PMCID: PMC8192220.

This one is a retrospective, single-center study of pediatric patients that used cell-free plasma next generation sequencing in their diagnostics. cfNGS was compared to standard of care diagnostic tests within 3 days of cfNGS testing. 110 patients were included, which totaled 142 cfNGS unique tests. The most common reason for ordering the tests (which could only be ordered via ID consultation) were signs of infection (48%) or focal imaging finding (27%). 105 tests were positive, of which 69 (66%) were considered clinically relevant. 32% of the results were felt to have changed patient management:

When compared to conventional diagnostic testing, the positive percent agreement of cfNGS was 89.6% while the negative percent agreement was 52.3%. So not terribly impressive, but a bit more data backing up the use of the Karius test. 

George MP, Esquer Garrigos Z, Vijayvargiya P, Anavekar NS, Luis SA, Wilson WR, Baddour LM, Sohail MR. Discriminative Ability and Reliability of Transesophageal Echocardiography in Characterizing Cases of Cardiac Device Lead Vegetations Versus Noninfectious Echodensities. Clin Infect Dis. 2021 Jun 1;72(11):1938-1943. doi: 10.1093/cid/ciaa472. PMID: 32533828.

Do you know how to tell apart a normal vegetation off an ICD lead from an infectious one? Probably not. What about a general cardiologist? Maybe not. But what about an echocardiologist?  This was a cohort study that evaluated 25 patients with CEID and positive blood cultures with TEE and compared them with non-infected controls who had a TEE and had an echodensity. 2 board-certified echocardiologists reviewed the TEE images in both groups. There were no difference in the size of the echodensity nor the mobility of it. Furthermore, both echocardiographers made a prediction of infection in 6/25 cases, with overall poor agreement of 68%. The sensitivity for infection ranged from 31.5% to 37.5%:

I think this should tell people that, just because you see an echodensity, in the absence of positive blood cultures or symptoms of infection, you should not go down the IE or device infection rabbit hole.