Symptoms of Infectious Diseases in Immunocompromised Travelers: A Prospective Study With Matched Controls
Gijs G. Baaten, MD, PhD, Department of Infectious Diseases, Public Health Service (GGD) Amsterdam, PO Box 2200, 1000 CE Amsterdam, The Netherlands. E-mail: firstname.lastname@example.org
Background. Immunocompromised travelers to developing countries are thought to have symptomatic infectious diseases more often and longer than non-immunocompromised travelers. Evidence for this is lacking. This study evaluates whether immunocompromised short-term travelers are at increased risk of diseases.
Methods. A prospective study was performed between October 2003 and May 2010 among adult travelers using immunosuppressive agents (ISA) and travelers with inflammatory bowel disease (IBD), with their non-immunocompromised travel companions serving as matched controls with comparable exposure to infection. Data on symptoms of infectious diseases were recorded by using a structured diary.
Results. Among 75 ISA, the incidence of travel-related diarrhea was 0.76 per person-month, and the number of symptomatic days 1.32 per month. For their 75 controls, figures were 0.66 and 1.50, respectively (p > 0.05). Among 71 IBD, the incidence was 1.19, and the number of symptomatic days was 2.48. For their 71 controls, figures were 0.73 and 1.31, respectively (p > 0.05). These differences also existed before travel. ISA had significantly more and longer travel-related signs of skin infection and IBD suffered more and longer from vomiting. As for other symptoms, no significant travel-related differences were found. Only 21% of immunocompromised travelers suffering from diarrhea used their stand-by antibiotics.
Conclusions. ISA and IBD did not have symptomatic infectious diseases more often or longer than non-immunocompromised travelers, except for signs of travel-related skin infection among ISA. Routine prescription of stand-by antibiotics for these immunocompromised travelers to areas with good health facilities is probably not more useful than for healthy travelers.
In recent years, international travel to developing countries has increased enormously.1,2 The number of travelers with a preexisting medical condition has probably also increased.3 This includes travelers using immunosuppressive agents (ISA), for example, because of a rheumatic disease, a solid-organ transplantation, or an auto-immune disease, and travelers with an inflammatory bowel disease (IBD). Due to better treatment options for these immunocompromised travelers, their overall health improves, and so does their motivation and physical fitness for travel. Indeed, the proportion of ISA and IBD among visitors of the travel clinic of the Public Health Service Amsterdam increased from 0.4% in 2001 to 0.9% in 2008. However, traveling to a developing country may complicate an underlying medical condition and may require special considerations and advice.4–6
Some travel health guidelines recommend that all travelers carry antibiotics for stand-by treatment. Yet, Dutch, British, and Canadian travel health guidelines recommend that only travelers with certain preexisting medical conditions, such as ISA or IBD, and travelers to areas with poor health facilities should be prescribed stand-by antibiotics for treatment of diarrhea.7–9 Although there is an abundance of recommendations on the immunocompromised traveler, data on the association of ISA or IBD with (tropical) infections, and on the benefits of preventive and therapeutic measures are lacking.
The number of ISA or IBD who visit developing countries is not known. In developed countries, the prevalence of rheumatic disease, psoriasis, or a solid-organ transplantation for which immunosuppressive agents are used is estimated at 0.7%;10,11 the prevalence of inflammatory bowel diseases is about 0.4%.12
To improve travel advice for this group, we conducted a prospective study with matched controls to see if ISA or IBD are more susceptible to travel-related symptomatic infectious diseases. We also studied the usage of antibiotics for stand-by treatment of diarrhea among these travelers.
A prospective study with matched controls was performed among travelers who attended the travel clinics of the Public Health Service Amsterdam or the University Medical Centre Leiden between October 2003 and May 2010. Both travel clinics provide residents of the cities of Amsterdam and Leiden with pre-travel health consultation and vaccinations according to Dutch travel health guidelines; visitors represent the general population of both cities.
All persons 18 years or older and (1) using immunosuppressive agents or (2) having an inflammatory bowel disease were eligible if planning to travel to one or more developing countries together with a non-immunocompromised travel companion, who was within 10 years of their own age. Thus, the control group was comparable for travel destination, travel duration, and exposure. Developing countries were defined as those with moderate to high risk on hepatitis A according to the World Health Organization.13
Immunosuppressive agents were defined as agents that completely or partly suppress one or more factors in the immune system, based on the classification of the WHO Collaborating Centre for Drug Statistics Methodology.14 For corticosteroids, only daily therapy with more than 10 mg of systemic prednisone per day or equivalent, for at least 2 weeks, was considered immunosuppressive, except when used as replacement therapy.15
Inflammatory bowel disease was defined as Crohn's disease or ulcerative colitis, diagnosed by a gastroenterologist.
Survey Methods and Definition of Symptoms
A standard questionnaire was used to collect data on socio-demographics and medical history. Items asked for were sex, age, country of birth, use of immunosuppressive agents, and history of inflammatory bowel disease.
Participants were asked to fill out a structured diary from the day they visited the travel clinic (up to 4 weeks before departure), until 2 weeks after return from travel. Recorded in the diary were travel itinerary; any episodes of fever, diarrhea, vomiting, rhinitis, cough, signs of skin infection, and fatigue; consultation with a doctor; and use of antibiotics or other medication. ISA pairs also recorded any episodes of arthralgia; IBD pairs recorded any episodes of abdominal pain. Fever was defined as a self-measured body temperature of 38.5°C or more. Diarrhea was defined as loose or watery stools. Rhinitis was defined as nasal discharge or congestion. Cough could be dry or productive. Signs of skin infection included redness, swelling, tenderness, and/or pus-like drainage. Arthralgia was defined as inflammatory or non-inflammatory joint pain. Abdominal pain could be acute or recurrent.
An episode of a symptomatic infection was defined as an aforementioned symptom at one or more consecutive days. Data were collected before departure to gain information about baseline symptoms, and for 2 weeks after return to encompass incubation periods of the most (acute) travel-related infectious diseases. In the Results section, the term “travel-related” refers to the period of travel itself and the 2 weeks thereafter.
According to the Dutch national guidelines on travel advice, of the pairs included, only the immunocompromised travelers were prescribed ciprofloxacin (500 mg 2 times a day, for 5 days in case of ISA and for 3 days in case of IBD), to be used as immediate self-treatment after the first passage of loose or watery stools.7 Controls were advised to see a doctor in case of diarrhea with fever, blood in stools, or diarrhea persisting for 3 days or more.7
Power analysis showed that 70 pairs were needed to prove a diarrhea outcome ratio of 2 or more, with α = 0.05and power = 80%.
This study was approved by a medical ethics committee. All participants gave their informed consent.
A random effects Poisson regression model was used to estimate incidence rates (IRs) and accompanying incidence rate ratios (IRR). IR was defined as the number of symptom onsets divided by the sum of symptom-free days for all individuals during a specific time period. A random effects logistic regression model was used to estimate the number of symptomatic days and accompanying odds ratios (ORs). The number of symptomatic days reflects an individual's probability to have a symptom on an arbitrary day. It was calculated to compare the disease burden between the immunocompromised travelers and their controls.
The random effects model takes into account two levels of correlation: (1) immunocompromised travelers and their travel companions had more or less the same exposure, and thus are not independent; (2) for IRs, there may be repeated episodes of a symptom within an individual; for numbers of symptomatic days, presence of symptoms over the days within an individual are correlated. ISA pairs and IBD pairs were analyzed separately.
For estimation of the parameters, a Bayesian approach was used, starting with non-informative priors. Posterior distributions were obtained by Markov Chain Monte Carlo methods, using the WinBUGS program.16,17 Three chains were generated, based on different sets of starting values. Parameter estimates are the medians of the posterior distributions. The range from the 2.5% to the 97.5% quantile is used to quantify the uncertainty in the parameter estimates. This range can be interpreted as a 95% confidence interval and will be referred to as such. If 1 is not included in the 95% confidence interval of a ratio, the ratio was considered statistically significant. When the incidence of a symptom in a group was zero, the approach as described in Firth was used,18 by means of the brglm package in R.19,20
During the study period, 99 ISA and 114 IBD, planning to travel with a non-immunocompromised travel companion, were eligible for inclusion. Of the ISA pairs, 16 (16%) did not want to participate and 8 (8%) were lost to follow-up after inclusion. Of the IBD pairs, 31 (27%) did not want to participate and 12 (11%) were lost to follow-up. The remaining participants all provided a completed diary.
Characteristics of the Study Sample
The study sample comprised 75 ISA and their 75 controls, and 71 IBD and their 71 controls. Of these 146 pairs, 124 (85%) were included at the Public Health Service Amsterdam and 22 (15%) at the University Medical Centre Leiden. Table 1 shows their characteristics.
Table 1. Characteristics of the cohort of travelers using immunosuppressive agents (ISA) and their controls, and of the cohort of travelers with inflammatory bowel disease (IBD) and their control
|Number of participants||75|| ||75|| ||71|| ||71|| |
|Median age in years*||48||(34–57)||46||(34–58)||35||(28–42)||34||(29–44)|
|Country of birth|
| Western country||67||89%||65||87%||63||89%||63||89%|
| Non-western country||8||11%||10||13%||8||11%||8||11%|
| Middle East and North Africa||13||17%||Idem|| ||13||18%||Idem|| |
| Sub-Saharan Africa||7||9%||Idem|| ||8||11%||Idem|| |
| Asia||46||61%||Idem|| ||32||45%||Idem|| |
| Latin America||9||12%||Idem|| ||18||25%||Idem|| |
|Median travel duration in days*||16||(13–22)||Idem|| ||16||(14–22)||Idem|| |
|Median duration data collection|
| Before departure (in days)*||9||(4–20)||Idem|| ||14||(4–20)||Idem|| |
| After departure (in days)*||30||(26–36)||Idem|| ||30||(27–36)||Idem|| |
| Rheumatic disease||51||68%||0|| ||0|| ||0|| |
| Psoriasis/Arthritis psoriatica||8||11%||0|| ||0|| ||0|| |
| Kidney transplantation||9||12%||0|| ||0|| ||0|| |
| Other||7||9%||0|| ||0|| ||0|| |
| Crohn's disease||0|| ||0|| ||37||52%||0|| |
| Ulcerative colitis||0|| ||0|| ||34||48%||0|| |
| No immunosuppressives||0||0%||75||100%||22||31%||71||100%|
| Systemic corticosteroids||21||28%||0|| ||2||3%||0|| |
| Methotrexate||45||60%||0|| ||1||1%||0|| |
| TNF-alpha inhibitor||19||25%||0|| ||4||6%||0|| |
| Oral mesalazine/sulfasalazine||4||5%||0|| ||37||52%||0|| |
| Other immunosuppressives||32||43%||0|| ||13||18%||0|| |
| Rectal immunosuppressives||0||0%||0|| ||12||17%||0|| |
Sixty-five ISA (86%) and 58 IBD pairs (82%) matched for country of birth. Only 10 ISA (13%) and 18 IBD pairs (25%) matched for gender.
The median travel duration was 16 days in both groups.
Of the ISA, 68% had a rheumatic disease. Of IBD, 52% had Crohn's disease and 48% had ulcerative colitis.
Of the ISA, 40 (53%) used one immunosuppressive agent, 24 (32%) two immunosuppressive agents, and 11 (15%) three immunosuppressive agents. Of IBD, 22 (31%) had not used any immunosuppressive agent, 30 (42%) used one immunosuppressive agent, 16 (23%) two immunosuppressive agents, and 3 (4%) three immunosuppressive agents.
IRs and Number of Symptomatic Days
Table 2 shows the travel-related symptoms by prevalence, IR, mean duration among symptomatics, and the number of symptomatic days per symptom for ISA and their travel companions. The figure in Table 2 shows the accompanying IRR and OR on a logarithmic scale. Likewise, Table 3 shows the results for IBD and their controls. Data concerning the occurrence of pre-travel-related symptoms are described in the text whenever relevant, and are not presented in the tables.
Travel-related symptoms of infectious diseases among travelers using immunosuppressive agents (ISA) and their non-immunosuppressed controls
Travel-related symptoms of infectious diseases among travelers with inflammatory bowel disease and their non-immune-suppressed controls
ISA and Controls
The prevalence of travel-related diarrhea was 47% among ISA and 40% among controls. The IR of travel-related diarrhea was 0.76 versus 0.66 per person-month; the IRR showed no significant difference. The number of days with diarrhea was 1.32 per month among ISA, comparable to controls. Also before travel, diarrhea outcome measures showed no significant differences between ISA and controls. For both ISA and controls, diarrhea outcome measures were significantly higher during travel than before travel.
The IR and the number of days for signs of skin infection were significantly higher among ISA than among controls, both before and during travel. Only among ISA, the outcome measures for signs of skin infection increased after departure.
The travel-related IR and number of days for fatigue and arthralgia were higher among ISA than among controls. However, these measures also differed before travel and showed no significant increase after departure.
The ISA and controls did not significantly differ in travel-related IRs and the number of symptomatic days for vomiting, fever, cough, and rhinitis, nor did they differ pre-travel. Although outcome measures for vomiting, fever, and rhinitis seemed to increase after departure, these differences were not significant.
IBD and Controls
The prevalence of travel-related diarrhea was 52% among IBD and 46% among controls. The IR was 1.19 per person-month versus 0.73, and the number of days with diarrhea was 2.48 per month versus 1.31, both not significantly different. As expected, pre-travel diarrhea outcome measures were significantly higher for IBD than controls, and significantly increased after departure.
The travel-related IR and the number of symptomatic days of vomiting were significantly higher for IBD than controls, whereas the pre-travel outcome measures for vomiting did not differ significantly between both groups.
Travel-related outcome measures for abdominal pain were significantly higher for IBD than controls. These measures also differed before travel and showed a non-significant increase after departure.
The travel-related number of days for signs of skin infection was higher among IBD than among controls; the IR was comparable. Before departure, none of the IBD or controls had signs of skin infection.
Pre-travel and travel-related IRs and the number of symptomatic days for fever, cough, rhinitis, and fatigue were comparable between both groups. For both IBD and controls, outcome measures for fever, cough, rhinitis, and fatigue did not increase significantly after departure.
Treatment and Doctor Consultation
Only 10 of 35 ISA with diarrhea (29%) and 5 of 37 IBD (14%) used the stand-by antibiotics according to study protocol. Another two ISA with diarrhea and one IBD with diarrhea used half of the daily-recommended dosage, and two IBD used only one tablet after which the complaints subsided. Twenty-three ISA of 35 with diarrhea (66%) and 31 of 37 IBD (84%) did not use the stand-by antibiotics at all. Effect of the use of antibiotics on the duration of diarrhea was unclear because of small numbers.
As to travel-related doctor consultations, both ISA (12%) and IBD (11%) were comparable to their controls (11 and 10%, respectively).
This study evaluates whether persons using immunosuppressive agents or having an inflammatory bowel disease are at increased risk for developing symptomatic infectious diseases when traveling to developing countries. Results concern short-term travelers.
Although we hypothesized that ISA would have symptoms more often and longer than non-immunocompromised travelers, no differences in travel-related diarrhea, vomiting, fever, cough, or rhinitis were found. The ISA had signs of fatigue and arthralgia more often than their controls, but this was unrelated to travel. Apparently, these symptoms of chronic (rheumatic) disease did not worsen during travel. Only the burden of signs of travel-related skin infection was higher among ISA than among controls. A higher risk of skin and soft tissue infection among patients using immunosuppressive agents has been reported before, in particular following anti-tumor necrosis factor (TNF) alpha drugs.21,22 The increased risk of infection may also be due to the immunomodulatory effects of rheumatic disease.23,24 Indeed, in our study, 8 of 10 ISA with travel-related signs of skin infection had a rheumatic disorder, of which 4 (50%) used anti-TNF alpha drugs, opposed to 13 of 43 (30%) ISA with a rheumatic disorder without travel-related skin infection (p = 0.31). Because bacterial skin infection can be life-threatening, especially for immunocompromised persons, stand-by antibiotics for this may be useful for areas where the availability of proper treatment is poor. This needs further investigation.
For IBD, the IR and the median number of days with diarrhea and abdominal pain were higher than among controls, both before and during travel. The incidence and burden of vomiting were higher among IBD, in particular during travel.
The same was true for the burden of signs of skin infection, not the incidence. No differences in travel-related fever, cough, rhinitis, and fatigue were found.
Our study design had distinctive strengths. Structurally specified data were obtained prospectively and on a daily basis. Data collection started before departure to gain insight into preexisting symptoms. It continued until 2 weeks after return from travel to encompass incubation periods of the most (acute) travel-related infectious diseases. With a travel companion serving as a matched control, situational specifics for the immunocompromised travelers and controls were comparable, which minimized any differences in exposure to infectious agents between the two groups. Both groups also matched for age and country of birth, but not for gender: both ISA and IBD were more often female. Yet, prospective studies on travel-related infectious diseases found no association of symptoms of infectious diseases and gender.25–28
The prevalence of ISA and IBD among visitors of the travel clinic of the Public Health Service Amsterdam in 2008 was 0.5 and 0.4%, respectively, comparable with the general population in a developed country.10–12 Also, the ages of our subjects with rheumatoid arthritis or IBD were comparable with the general population. Participants' travel destinations were equally distributed across the four regions. Their median travel duration of 20 days corresponded well with the median travel duration of the average traveler.29,30 Thus, the study sample can be considered representative, and results can reasonably be applied to the average traveler with ISA or IBD to a developing country. Nevertheless, our findings represent immunocompromised persons who were well and confident enough to travel.
This study also had some limitations. Sample size may not have been large enough to detect small differences. Secondly, some of the symptomatic illnesses could have been due to a non-infectious cause. This may have overestimated the (absolute) rate of infection in all groups. In particular for IBD, recognizing the difference between travel-related diarrhea versus an exacerbation of their disease may have been difficult. Thirdly, although the diary provided information on symptom duration, it did not distinguish mild symptomatology from severe. For example, immunocompromised travelers could have had more bowel movements or more water loss. Finally, the immunocompromised travelers and controls differed in counseling and prescription, and some immunocompromised travelers did use the stand-by antibiotics. Therefore, the data may be skewed toward seeing fewer differences in outcome measures between both groups.
Our findings represent immunocompromised persons and their travel companions who sought pre-travel health advice. They may have had a more than average health awareness, particularly having received travel advice and knowing the objectives of the study. As to usage of stand-by antibiotics, its importance was emphasized by an experienced travel health expert, and by means of information leaflets. Nevertheless, 66% of ISA with travel-related diarrhea and 84% of IBD with travel-related diarrhea did not use this treatment. Of 146 stand-by antibiotic courses provided, 131 (90%) were not used.
Although studies have shown that immunocompromised persons are at increased risk of severe outcome for some infectious diseases, including food- and waterborne infections,31–33 the increased risk of gastroenteritis among ISA has not been firmly established in controlled studies,21,23 nor in our study. For IBD, factors that predispose to infectious complications are the disease process itself and the use of immunosuppressive medication.34 Unfortunately, these factors could not be addressed in our study because of small numbers. Nevertheless, in our study, the higher IR and number of days of diarrhea among IBD as compared to controls appeared to be unrelated to travel. Thus, routine prescription of stand-by antibiotics for uncomplicated diarrhea for ISA or IBD is probably not more useful than for healthy travelers. Stand-by antibiotics may be useful for immunocompromised travelers to areas where health facilities are lacking in case of more severe illness, for example three or more unformed stools per 24 hours with accompanying symptoms such as fever, or blood in stools. The merits of this definition could not be assessed in this study.
In conclusion, in this study, short-term travelers using immunosuppressive agents or having an inflammatory bowel disease did not have travel-related symptoms of diarrhea, fever, cough, rhinitis, fatigue, and arthralgia more often or longer than non-immunocompromised short-term travelers. Among ISA, the incidence and burden of signs of travel-related skin infection were higher. Among IBD, the incidence and burden of vomiting were higher.
Our findings indicate that routine prescription of stand-by antibiotics for these immunocompromised travelers to areas with good health facilities is probably not more useful than for healthy travelers. More prospective studies are needed.
The authors thank the nurses and medical doctors of the Public Health Service Amsterdam and the University Medical Centre Leiden for their assistance in subject inclusion and data collection, and Roel A. Coutinho for his critical review of the manuscript. This study was financially supported by grant 7115 0001 from ZonMw, the Netherlands Organization for Health Research and Development.
Declaration of Interests
The authors declare that they have no conflicts of interest.