Psychiatric morbidity after surgery for inflammatory bowel disease: A systematic review

Table 1 Characteristics of included studies

Ref.	Year	Country	Study design	Population	Surgery	Outcome measures	Comparisons
Makkar et al[22]	2015	Canada	Cross sectional study	137 patients with UC > 18 yr who were > 1 yr from the final stage of their total IPAA surgery.	IPAA	DASS-21 including subscales for stress, anxiety and depression	Subgroup analysis comparing normal pouch, irritable pouch syndrome and pouch inflammation. All groups had IPAA
Panara et al[4]	2014	United States	Retrospective cohort study	393 patients > 18 yr with UC (121) or CD (272)	History of surgical stoma or seton placement as risk factor (from surgical records)	ICD-9-CM (International Classification of Diseases, Clinical Modification) codes for depression	None
Ananthakrishnan et al[16]	2013	United States	Retrospective cohort study	707 with CD and 530 with UC	Bowel resection surgery (ICD records)	ICD-9 codes for depressive disorders or generalized anxiety given after 30days after surgery.	IBD patients not having surgery and patients undergoing surgery for other diseases
						Analyses of independent predictors of depression and anxiety following IBD-surgery
Knowles et al[14]	2013	Australia	Cross sectional study	83 mixed IBD. (62.7% UC) Age between 18-40 yr	Stoma surgery (self-reported)	HADS (normal = 0-7, mild severity = 8-10, moderate severity = 11-15, severe severity = 16-21)	none
Knowles et al[15]	2013	Australia	Cross sectional study	31 with CD	ostomy	HADS	none
						(normal = 0-7, mild severity = 8-10, moderate severity = 11-15, severe severity = 16-21)
Nahon et al[3]	2012	France	Cross sectional study	1663 with IBD (63.9% CD and 37.1% UC or indeterminate colitis	Past history of surgery as risk factor	HADS > 11 on either subscale was considered “significant” cases of psychological comorbidity	none
Schmidt et al[21]	2007	Germany	Cross sectional study	43 with UC	IPAA	HADS ≥ 11 on either subscale (depression/anxiety) indicative of a probable mental disorder	IPAA patients in remission, with pouchitis and with irritable pouch syndrome
Häuser et al[20]	2005	Germany	Cross sectional study	101 with UC	IPAA	HADS ≥ 11 on either subscale was considered “significant” cases of psychological comorbidity	UC patients with IPAA vs general german population and UC patients with IPAA vs UC patients without IPAA.
						Use of psychopharmacological agents
de Oca et al[23]	2003	Spain	Cross sectional study	100 with UC and 12 with CD (discovered postoperative)	IPAA	STAI for Anxiety	Only subgroup (CD vs UC) comparisons
Nordin et al[19]	2002	Sweden	Cross sectional study	331 with UC and 161 with CD (all in the range of 18-70 yr of age)	Ileostomy, ileoanal anastomosis and ileorectal anastomosis	HADS where ≤ 7 = “non-case”; 8-10 = “doubtful case”; ≥ 11 = “case”	none
Tillinger et al[18]	1999	Austria	Prospective cohort study	16 with CD	Elective ileum or colon resection	Beck depression inventory within one week before operation, three, six and 24 mo postoperative	none
Keltikangas-Järvinen et al[17]	1983	Finland	Cross sectional study	32 with UC operated with ileostomy	Operation with ileostomy (follow up = 7 ± 1.2 yr. after the operation)	Beck’s depression scale and Rorschach content interpretation for anxiety	34 colorectal cancer patients having colostomy

Table 2 Results regarding depression and anxiety

Ref.	Depression results	Anxiety results
Nahon et al[3], 2012	Multivariate analysis of predictive factors found no association between past history of surgery and depression (OR = 0.93, 95%CI: 0.50-1.72)	Multivariate analysis of predictive factors found past history of surgery to be significantly associated with decreased risk of anxiety (OR = 0.47, 95%CI: 0.31-0.71)
Panara et al[4], 2014	Multivariate analysis: history of surgery had a non-significant HR = 1.3 (95%CI: 0.92-1.76; P = 0.13).	-
Ananthakrishnan et al[16], 2013	Chi-square test: Higher 5 yr postoperative risk in IBD group (16%) compared with diverticulitis (9%) and inguinal hernia group (7%) (P < 0.05). Higher risk in CD surgery group compared with non-surgical group (OR = 1.34, 95%CI: 1.01-1.77). No significant increased risk in UC surgery group compared with non-surgical group (OR = 1.21, 95%CI: 0.93-1.58).	no significant increased OR in CD-surgery group compared with non-surgical group (OR = 1.20, 95%CI: 0.93-1.55) or UC-surgery group compared with non-surgical group (OR = 1.26, 95%CI: 0.96-1.65).
Keltingas-Jarvinen et al[17], 1983	Comparisons of means in Beck depression inventory – type of analysis not stated: UC < colorectal cancer	Comparisons of means in Rorschach content interpretation for anxiety – type of analysis not stated: UC > colorectal cancer
Tillinger et al[18], 1999	Wilcoxon test: significantly improved score three and six months postoperatively (P = 0.0038 and 0.0013 respectively). 24 mo postoperatively only improved scores for patients still in remission.	-
Nordin et al[19], 2002	Percentage of population divided on HADS depression subscales: 87% “non-cases”; 9% “doubtful cases”; 4% cases	Percentage of population divided on HADS anxiety subscale: 71% “non-cases”; 14% “doubtful cases”; 15% cases.
	Subgroup analysis of depression: unpaired t-test showed no difference between CD and UC patients with ileostomies and those without ileostomies.	Subgroup analysis of anxiety: unpaired t-test showed no difference between CD and UC patients with ileostomies and those without ileostomies.
Knowles et al[14], 2013	Percentage of population divided on HADS depression subscales: 84% normal; 6% mild; 10% moderate; 0% severe	Percentage of population divided on HADS anxiety subscale: 50% normal; 24% mild; 16% moderate; 10% severe.
Knowles et al[15], 2013	Percentage of population divided on HADS depression subscales: 58% normal; 26% mild; 16% moderate-severe	Percentage of population divided on HADS anxiety subscale: 51% normal; 39% mild; 10% moderate-severe
Häuser et al[20], 2005	Student’s t-test: no increased probable (HADS ≥ 11) mental disorder in UC with IPAA vs the general German population. Wilcoxon Mann-Whitney test: no difference in HADS depression subscales between UC patients with IPAA† compared to UC without IPAA.	Student’s t-test: no increased probable (HADS ≥ 11) mental disorder in UC with IPAA vs the general German population. Wilcoxon Mann-Whitney test: no difference on HADS anxiety subscale between UC patients with IPAA compared to UC without IPAA.
Schmidt et al[21], 2007	Kruskal-Wallis test showed no significant difference in HADS depression subscales between IPAA subgroups	Kruskal-Wallis test showed no significant difference on HADS anxiety subscale between IPAA† subgroups
Makkar et al[22], 2015	ANOVA: Significant difference between DASS among patients with irritable pouch syndrome (11.7 ± 9.7), pouch inflammation (8.1 ± 9.1) and normal pouch (4.4 ± 6.2), P =0.012.	ANOVA: no significant difference between DASS among patients with irritable pouch syndrome (8.1 ± 7.0), pouch inflammation (6.0 ± 6.8), and normal pouch (4.2 ± 4.9), P = 0.1
de Oca et al[23], 2003	-	Student’s t-test: CD < UC on anxiety values of the STAI ( P = 0.014)

OR: Odds ratio; ANOVA: Analyses of variance; HADS: Hospital Anxiety and Depression Scale; DASS: Depression, Anxiety and Stress Scale; STAI: State-Trait Anxiety Inventory; IPAA: Ileal Pouch-Anal Anastomosis; CD: Crohn’s disease; UC: Ulcerative colitis.