Unanswered Questions on the Management of Nocardia Infections in Transplant Recipients

Several questions remain open on the subject of nocardiosis management, and, in some cases, recommended practices are known but not implemented for various reasons (even the level of evidence to support recommendation). The choice of initial antibiotic treatment (before species identification), the use of a monotherapy as opposed to multidrug regime, and the ideal duration of treatment are all choices that depend mainly on the disease’s clinical presentation in each individual case. It has been suggested that antibiotic duration can be shortened in selected patients, provided that central nervous system (CNS) involvement is excluded. However, ideally, individualized treatment plans should take into consideration the following factors: (a) species identification or susceptibility testing (ST), (b) assessment of invasive disease and CNS involvement, (c) response to treatment, and (d) adverse effects.

Nocardiosis is an uncommon opportunistic infection caused by gram-positive, slow-growing aerobic actinomycetes of the genus Nocardia. Disease occurs predominantly in immunosuppressed patients.¹ After organ transplantation, the incidence of nocardiosis varies according to the transplanted organ, usually ranging from <1% after kidney or liver transplantation to 1%–3.5% after heart or lung transplantation. The risk of developing nocardiosis is highest in the first year following the procedure.²

According to studies of transplant recipients, risk factors for nocardiosis include augmented immunosuppression, cytomegalovirus disease, and patient age.^3,4 Regarding outcomes, however, there is less information available. Concerning mortality outcomes, it is important to note the difficulty of obtaining high-quality data. Given the rarity of nocardiosis in solid organ transplant (SOT) recipients, the available studies tend to be retrospective and to present smaller sample sizes. Mortality rates from nocardiosis can vary widely based on the characteristics of the sample population, as well as on the duration of follow-up. Factors that may influence mortality outcomes include clinical presentation, presence of co-infections and comorbidities, species susceptibility, treatment options, and duration.^5,6

In this issue of Transplantation, Yetmar et al⁷ attempt to shed further light on these questions. In their multicenter, retrospective cohort study, conducted at 3 Mayo Clinic sites in Arizona, Florida, and Minnesota, the authors analyzed a series of factors associated with 1-y mortality in 125 recipients of SOT recipients diagnosed with culture-confirmed nocardiosis, over the period of 2000–2020. The overall mortality rate of 16.8% is consistent with similar studies on the subject. Recipients of liver transplants presented the highest mortality rate of all SOT recipients. Liver transplantation (hazard ratio [HR] 3.52; 95% confidence interval [CI] 1.27-9.76) and time from symptom onset to presentation (HR 0.92 per day; 95% CI 0.86-0.99) were independently associated with 12-mo mortality, while disseminated infection was not (HR 1.23; 95% CI 0.49-3.13).⁷ The higher mortality for liver recipients stands out as an interesting finding, and although the small number of recipients of this type limits the extent to which conclusions may be drawn, this result may nonetheless provide a productive avenue for future research.

Nocardia frequently affects the lungs, brain, and skin, and disseminated infection is not uncommon in transplant recipients.^4,6 The sites of nocardial infection may possibly correlate with poorer outcomes. In this context, brain imaging is highly recommended, even in patients with no neurological symptoms, since CNS nocardiosis was identified in around 40% of SOT patients presenting no specific symptoms.³ If there is CNS involvement, treatment is maintained for longer periods.⁵ In this article, only 14% of patients were reported as presenting CNS disease. However, given that only two-thirds of the total number of patients underwent imaging for CNS, it is possible that the prevalence was under-reported.

Many different Nocardia species are able to cause disease in humans, and the predominance of a particular species will vary according to geographical region. Molecular methods have led to the reclassification and renaming of several Nocardia isolates, and, as each species may present a specific susceptibility pattern, species identification and antimicrobial ST are therefore strongly recommended.⁸ In the present study,⁷ the most common species identified were Nocardia farcinica (22.4%), Nocardia cyriacigeorgica (15.5%), and Nocardia nova (15.5%). Susceptibility to trimethoprim-sulfamethoxazole (TMP-SMX) and linezolid was high across all species studied, whereas a significant percentage of isolates showed resistance to beta-lactams. In previous studies, Nocardia farcinica was more frequently associated with brain, skin, and soft tissue infections than were other species, and was also shown to be resistant to B-lactam antibiotics in most cases.³

All patients with disseminated disease underwent ST and there were no differences observed in outcome based on the use of >1 active drug in the first 4 wk of therapy. It should be noted that this cohort presented a low rate of monotherapy but a high rate of TMP-SMX use.⁷ TMP-SMX was active in >90% of isolates and patients presented good response to treatment, a fact in line with previous results in the literature.⁹ Linezolid was active against all Nocardia isolates and survivors had a higher rate of linezolid use, although this analysis was likely underpowered. Overall, results suggest that multiple agents may only be necessary before the availability of ST results.

In the present study, disseminated infection was not independently associated with poor outcomes and, as mentioned previously, specific sites within patients with disseminated disease may be more predictive. In patients having survived the initial 12-mo period postdiagnosis, average length of treatment was approximately 300 d. A total of 26.6% of subjects presented disseminated disease, and 14% presented disease in the CNS. Both these scenarios are traditionally associated with longer treatment times and drugs used in combination. Reported results are consistent with some previous studies on the subject, and there appears to be a paradigm shift regarding outcome prediction indicators such as disseminated disease, and combination therapy.^5,10,11 These indicators and their correlation with patient outcomes will require further investigation.

In this issue of the journal,⁷ authors present information concerning 1-y mortality in SOT recipients with nocardiosis, as well as some treatment-specific issues. Undoubtedly, this study brings new and important data to assess the risk of mortality and treatment for disseminated nocardiosis in the SOT population. Limitations include its retrospective design and lack of complete data regarding CNS involvement.

The rarity of this combination of events (namely, of nocardiosis following SOT) restricts large, randomized, prospective research of the type that would be needed to address queries related to underlying immunosuppression, initial antibiotic treatment, and duration. No study should be interpreted in isolation and although some studies, such as the present, provide more information about the management and treatment of nocardiosis in transplant patients, a more precise assessment on treatment strategies based on specific Nocardia syndromes and outcome in the setting of organ transplantation is still lacking.