(See the Major Article by Echenique et al on pages 729–37.)
Up to 27% of recipients of solid organ transplant (SOT) experience diarrhea posttransplant,
with some episodes being severe enough to require hospitalization [1–3]. Immunosuppressive
medications are a common cause, but reduction of immunosuppression carries the risk
of allograft rejection, making it necessary to rule out other causes, such as infection,
before modifying the immunosuppressive regimen [3–7].
In this issue of Clinical Infectious Diseases, Echenique and colleagues report the
frequency of infectious causes and diagnostic yield of microbiologic tests in a large
retrospective cohort of hospitalized SOT patients with community- and hospital-onset
diarrhea [8]. An implied goal of the study was to determine if a targeted approach
to testing for infectious causes of diarrhea could be considered as an alternative
to the more comprehensive testing strategies recommended at their facility and elsewhere
[7, 8]. With this question in mind, it is striking that most diarrheal episodes in
their study had no specific etiology identified, infectious or otherwise, and >90%
resolved in the hospital before discharge. A potential infectious cause was identified
in 30% of community-onset and 19% of hospital-onset diarrheal episodes overall. Most
infections were limited to just 3 agents—Clostridium difficile, norovirus, or cytomegalovirus—which
comprised >90% of the infectious cases identified. Other infectious causes of diarrhea,
such as Campylobacter, Salmonella, rotavirus, adenovirus, astrovirus, Cryptosporidium,
and Giardia, were rarely identified. In light of these findings, the authors suggest
that initial testing for infectious causes of diarrhea in SOT recipients could potentially
be limited to a few tests (eg, blood quantitative cytomegalovirus polymerase chain
reaction [PCR], C. difficile PCR, norovirus reverse transcription PCR, bacterial stool
culture) in most patients. Then, if symptoms persist, additional testing could be
performed to diagnose or rule out less common infectious causes before modifying immunosuppressive
regimens. An analogous tiered approach to testing has been used for decades in immunocompetent
patients with diarrhea.
However, there are several important points that should be considered before generalizing
the results of this study to other centers. First, the frequencies of several potentially
significant pathogens in immunocompromised hosts, such as Campylobacter and Cryptosporidium,
were lower in this study than previous studies based in Europe (Table 1). Conversely,
the rate of C. difficile among community-onset diarrheal episodes was higher than
other studies and comparable to the rate of C. difficile in hospital-onset diarrheal
episodes. This may be a manifestation of the changing epidemiology of C. difficile
with more transmission and infection in the community or simply due to community onset
of symptoms after hospitalization [11, 12]. Alternatively, it is possible that some
of the C. difficile identified in community-onset diarrheal episodes were recurrent
infections or lingering colonization with another cause of symptoms, as many patients
had a history of prior C. difficile infection. In any case, the differences in the
rates of individual diarrheal pathogens in this study vs previous studies suggests
that the etiologies of infectious diarrhea in SOT recipients likely differ between
transplant centers. Hence, the authors note in their discussion that the results of
this study should not be generalized without verification of the local epidemiology
and validation.
Table 1.
Frequency of Infectious Causes in Patients With Posttransplant Diarrhea, by Study
Study
Echenique, 2014 (This Issue) [8]
Maes et al, 2006 [7]
Roos-Weil et al, 2011 [9]
Coste et al, 2013 [10]
Test Methods
Molecular and Classical
Classical
Molecular and Classical
Multiplex Molecular Panels
Country
United States
Belgium
France
France
Population
CO Diarrhea
HO Diarrhea
Diarrhea
CO Diarrhea
CO Diarrhea
No Diarrhea
Transplant type
Mixed
Mixed
Renal
Renal
Renal
Renal
Positive (total)
30%
19%
28%
42%
56%–72%
20%–47%
Bacteria
Clostridium difficile
13%
12%
2%
10%
2%
0%–3%
Campylobacter, Salmonella
1%
0%
12%
4%
30%
3%–7%
Viruses
CMV
6%
3%
7%
0%–2%
0%–2%
NT
Norovirus
9%
4%
NT
17%
26%
10%–13%
Other GI viruses
0%
0%
1%
1%
6%
0%
Parasites
Cryptosporidium, Giardia
<1%
0%
2%
4%
4%
0%
Other
0%–<1%
0%
2%
3%
0%
0%–3%
Abbreviations: CMV, cytomegalovirus; CO, community-onset; GI, gastrointestinal; HO,
hospital-onset; NT, not tested.
A more concerning explanation for the low frequency of traditional infectious causes
of community-onset diarrhea in this study is underdiagnosis due to lack of testing.
Most diarrheal episodes in this study included an abbreviated microbiologic workup
in spite of a more complete institutional test protocol. Bacterial stool culture was
performed in only 58% of community-onset diarrheal episodes, and testing for Cryptosporidium
and Giardia was performed in only 27% of episodes, making it possible that infectious
cases were missed due to lack of testing. Less than 1% of episodes had testing for
gastrointestinal viruses other than norovirus. This is not the authors’ fault, as
the study was a retrospective summary of real-world practice and not a prospective
study designed to comprehensively define the infectious etiologies of diarrhea in
SOT recipients. The testing strategies observed in this study may also have been reasonable
and cost-effective medical practice because most diarrheal episodes appear to have
had a relatively short duration. However, it is important to be aware of the limited
workup for intestinal pathogens when interpreting the results of this study.
The final point that should be made is that diagnostic test methods for gastrointestinal
pathogens are rapidly and dramatically changing. Highly sensitive, multipathogen nucleic
acid amplification test panels are poised to become the routine diagnostic test for
gastrointestinal pathogens within the next few years. Two molecular test panels, each
detecting multiple diarrheal pathogens, are already approved for in vitro diagnostic
testing by the US Food and Drug Administration. Several other multiplex gastrointestinal
pathogen panels are in various stages of clinical trials and regulatory review. Similar
to what happened with the introduction of multiplex viral respiratory panels, it is
likely that these multiplex gastrointestinal pathogen panels will replace traditional,
less sensitive tests such as bacterial stool culture and viral and protozoal immunoassays
in the near future. Moreover, whether it is cost effective or not, these multiplex
molecular panels may obviate the need or choice to target testing to selected pathogens.
Investigators in France recently demonstrated the dramatic potential of these panels
to increase the detection of gastrointestinal pathogens in SOT recipients, leading
them and others to question the dogma that most posttransplant diarrhea is noninfectious
[10]. In this pilot study, the investigators retested stool from 54 episodes of severe
diarrhea in recipients of SOT with several multiplex gastrointestinal pathogen panels
and compared results with classical test methods (ie, culture, microscopic examinations,
immunoassays, some molecular tests) [10]. Strikingly, the proportion of samples with
1 or more potential pathogens detected increased from 23% with classical tests to
72% after performance of the multiplex gastrointestinal pathogen panels. Similar results
have been observed in nontransplant populations [13, 14]. However, it is worth noting
that asymptomatic transplant patients can also have pathogens detected by these multiplex
panels, making it essential to limit testing to symptomatic patients and correlate
test results clinically [10]. A partial summary of results from this study using multiplex
gastrointestinal pathogen panels are included in Table 1 (sixth and seventh columns)
for comparison with the Echenique et al study [8] and previous studies published elsewhere
[7, 9–10].
In summary, Echenique and colleagues provide valuable real-world data showing that
the majority of diarrheal episodes in SOT recipients have no etiology identified,
with infectious causes being limited to a few pathogens under existing test methods.
Looking forward, clinical implementation of multiplex gastrointestinal pathogen test
panels is likely to increase the proportion of diarrheal patients with a potential
pathogen identified, changing the question from which test(s) to order, to when to
test and when to treat.