Gastrointestinal Symptoms in Diabetes: Prevalence, Assessment
Total Page:16
File Type:pdf, Size:1020Kb
Diabetes Care Volume 41, March 2018 627 Yang T. Du,1,2 Christopher K. Rayner,2,3,4 Gastrointestinal Symptoms in Karen L. Jones,1,2,3 Nicholas J. Talley,5,6,7 Diabetes: Prevalence, Assessment, and Michael Horowitz1,2,3 Pathogenesis, and Management Diabetes Care 2018;41:627–637 | https://doi.org/10.2337/dc17-1536 If you haven’t measured something, you really don’t know much about it. —Karl Pearson (attributed) Gastrointestinal (GI) symptoms represent an important and often unappreciated cause of morbidity in diabetes, although the significance of this burden across the spectrum of patients and the underlying pathophysiology, including the relationship of symptoms with glycemic control, remain poorly defined. The relevance of GI symptoms and the necessity for their accurate assessment have increased with the greater focus on the gut as a therapeutic target for glucose lowering. This review addresses the prevalence, assessment, pathogenesis, and management of GI symp- toms in diabetes, beginning with broad principles and then focusing on specific seg- ments of the GI tract. We initially performed a literature search of PubMed by using REVIEW synonyms and combinations of the following search terms: “gastrointestinal symp- toms”, “diabetes”, “prevalence”, “pathogenesis”, “diagnosis”,and“management”. We restricted the search results to English only. Review papers and meta-analyses are presented as the highest level of evidence where possible followed by random- ized controlled trials, uncontrolled trials, retrospective and observational data, and expert opinion. 1Endocrine and Metabolic Unit, Royal Adelaide Hospital, Adelaide, South Australia, Australia 2Discipline of Medicine, The University of Ade- laide, Adelaide, South Australia, Australia PREVALENCE AND SIGNIFICANCE OF GASTROINTESTINAL SYMPTOMS IN 3National Health and Medical Research Council DIABETES Centre of Research Excellence in Translating Nu- tritional Science to Good Health, The University Although gastrointestinal (GI) symptoms generally are accepted as more common in of Adelaide, Adelaide, South Australia, Australia people with diabetes than in the general population, the reported prevalence has 4Department of Gastroenterology and Hepatol- varied substantially, being much higher ($70%) in most but not all outpatient samples ogy, Royal Adelaide Hospital, Adelaide, South (1–5) compared with community studies (6–11) (Table 1). These inconsistencies prob- Australia, Australia 5 ably reflect differences in the patient populations and the methodology used to eval- Faculty of Health and Medicine, University of New- castle, Newcastle, New South Wales, Australia uate symptoms. Whether symptom prevalence differs substantially between type 1 6Division of Gastroenterology and Hepatology, and type 2 diabetes is uncertain. In an Australian community study, GI symptoms Mayo Clinic, Rochester, MN tended to be less common in the former, but the number of patients with type 1 7Karolinska Institute, Stockholm, Sweden diabetes was small (9). In contrast, patients with type 1 diabetes in a U.S. commu- Corresponding author: Michael Horowitz, michael nity study experienced less heartburn but more constipation than those with type 2 [email protected]. diabetes (7). Received 26 July 2017 and accepted 7 December A high prevalence of GI symptoms exists in the general population, which may be 2017. influenced by BMI, sex, psychological comorbidities, Helicobacter pylori infection, and © 2018 by the American Diabetes Association. age (12). For example, 7–30% of adults in the community have constipation, and 7–10% Readers may use this article as long as the work is properly cited, the use is educational and not suffer from bloating (13). GI symptoms, particularly those deemed embarrassing (e.g., for profit, and the work is not altered. More infor- fecal incontinence), often are not reported unless patients are specifically questioned mation is available at http://www.diabetesjournals (13). In a community-based study of 777 Australian adults, obesity was independently .org/content/license. 628 GI Symptoms in Diabetes Table 1—Reported prevalence of GI symptoms in diabetes Prevalence (%) Studies showing a Symptom Community studies Tertiary center studies Community controls Tertiary center controls significant difference Esophageal Dysphagia 6.1I 5.4IV 5.9V 4.0A 12.9B 1.2I 1.7IV 4.1V 0A 7.8B IV Reflux/heartburn ;24I 11.6II (type 1) 19.8II (type 2) 13.5IV 19V 8.1A ;32B 44C ;5D 58.8E 10–23I 23II 11IV 18V 0A 22B 21C 4D I,IV,A,B,C,E Gastric Abdominal pain or discomfort 8.5–12.1I 15.1III 13.5IV 19.3V 16.1A 12.8–15.2C 15.1D 12.2–21.4I 12.6III 10.8IV 14.6V 4.9A 2.2–4.5C 10.3D I,IV,V,A,C Early satiety 26.8I 32.2III 5.2IV 6.7A 55.1B 12.5D 6.1I 20.2III 4.3IV 1.2A 49.6B 5.4D I,III,IV,A,D Postprandial fullness 18.6I 8.6IV 16.8A 8.5I 5.2IV 1.2A I,IV,A Bloating/abdominal distension 42.3I 21.0III 12.3IV 29.8V 21.5A 57.8B 19.6D 24.4I 15.2III 11.4IV 26.5V 13.4A 55.0B 13.3D I,III,IV,D Nausea 22.7I 16.8III 11.6II (type 1) 6II (type 2) 5.2IV 11.9V 3.4A 45.2B 4.3D 9.1I 5.5–10.6II 12.7III 3.5IV 5.7V 1.2A 32.1B 2.6D I,III,IV,V,B Vomiting 12.2I 5.6III 1.7IV 3V 9.6B 3I 4.3III 1.1IV 2.8V 3.4B I,IV,B Small and large intestines Diarrhea 18.6I 0II 15.6IV 22VI 18.9V 34.9A 41.0B 12.8C 17.9D 13.4I 0II 10.0IV 11.4V 9VI 4.9A 34.9B 2.2C 11.7D IV,VI,A,C,D Constipation 14.3I 16.7II (type 1) 10.1II (type 2) 11.4IV 27.5A 33.7B 16.1D 10.3I 11.5–13.5II 9.2IV 7.3A 32.1B 14.6D IV,A Fecal incontinence 0.7II (type 1) 4.6II (type 2) 2.6IV 9.9V 3.4A 8.8C 1.2–1.8II 0.8IV 4.6V 0A 0C IV,V,C Only English-language articles with full text available from PubMed; included matched controls and reported crude prevalence rates are shown. ISchvarcz et al. (6): community study, questionnaire not validated, 110 patients with long-standing type 1 diabetes, 210 controls. IIMaleki et al. (7): community study, used validated BDS, 138 patients with type 1 diabetes and 170 controls, 217 patients with type 2 diabetes and 218 controls. IIIRicci et al. (8): community study, face-to-face interview, 483 patients with diabetes, 422 controls. IVBytzer et al. (9): community study, validated questionnaire that was based on the BDQ, 423 patients with type 1 diabetes (5.2%) and type 2 diabetes (94.8%), 8,185 controls. VIcks et al. (10): community study, interview questions derived from Talley et al. (1992) questionnaire (14), 544 patients with type 2 diabetes, 544 controls. VIQuan et al. (11): community study, used validated DBSQ, 51 patients with type 1 diabetes, 128 patients with type 2 diabetes, 65 controls. AKo et al. (1): tertiary referral center, interview questions derived from Horowitz et al. (1989) questionnaire (15), 149 patients with type 2 diabetes, 82 controls. BMjornheim¨ et al. (2): tertiary referral center, questionnaire from Ruth et al. (1991) (16), 364 patients with type 1 diabetes, 242 controls, high crude prevalence rates probably because patients who answered mild, moderate, or severe to any question were considered to have the symptom. CAbid et al. (3): tertiary referral center, questionnaire from Talley et al. (32), 250 patients with type 2 diabetes, 264 controls. Dde Kort et al. (4): tertiary referral center, Gastrointestinal Symptom Rating Scale and PAGI-SYM questionnaires used, 280 patients with type 1 (28.9%) and type 2 (71.1%) diabetes, 355 controls, prevalence rates for clinically relevant GI symptoms (Gastrointestinal Symptom Rating Scale score $3, PAGI-SYM score $2). EHa et al. (5): tertiary referral center, GERD symptoms evaluated by using Frequency Scale of the Symptoms of GERD questionnaire, 258 patients with type 2 diabetes, 184 controls. in adherence to therapy (22). wasassociatedwithan rence of GI symptomspatients with with type 2 metformin diabetes, thediabetes medications. In occur- newly diagnosed symptoms also may affect tolerabilityhas of not been evaluatedGI in symptom improvement diabetes. (13), GI butrelated this quality of life isGI concordant with disorders, improvement in health- to be reduced by paresis, annual income has been reported tients with symptomatic diabetic gastro- GI symptom groups increases (20). In pa- crease markedly as the number of distinct Scores on all Short Form 36 subscalesabetes de- negatively and substantially (20). turnover (11). vincingly associated with symptom nomic neuropathy have not beenthem con- (11). Glycemicresolution, control with and a auto- twofoldfold risk risk of of losing gaining GIsion symptoms was and associated its with aboutand type a 2 three- diabetes, the onset ofcommunity-based depres- patients with typelence remains 1 relatively constant (11). In ance of others so thatbe the overall counterbalanced preva- by thethe onset disappear- of new“ symptoms appears to tes, there isbased substantial subjects symptom with anddiabetes without is poorly diabe- de remains unclear. and/or represents the outcome ofchological them distress causesThe symptoms fundamental issue ofin whether those psy- with anxiety orsymptoms depression were (19). about twice as frequent predominantly with typein 2 a community diabetes, study of ciated with GI symptoms.and For depression (9,19), example, are strongly asso- logical comorbidities, includingchosocial anxiety distress than men (9);women, psycho- who exhibit higher levels of psy- symptoms occur more frequently in prevalence of GI symptoms in diabetes. is essential in studiesinclusion of an related appropriate control to group the higher in women (18).alence Accordingly, of the functional GIin disorders control populations also (6,9), and the is prev- derance of GI symptoms exists in females creased risk of heartburn (17).