Association of In-Hospital Surgical Bleeding Events with Prolonged Hospital Length of Stay, Days Spent in Critical Care, Complications, and Mortality: A Retrospective Cohort Study Among Patients Undergoing Neoplasm-Directed Surgeries in English Hospitals

Figures & data

Figure 1 Crude incidence proportion of bleeding events during index admission*.

Notes: *The proportion of bleeding events first documented during the surgical episode was: 96.5% low anterior resection, 93.9% hysterectomy, 81.5% lung surgery, 97.9% mastectomy, and 100.0% prostate surgery; the proportion of patients with bleeding events who also had reoperation for surgical arrest of bleeding was: 0.0% low anterior resection, 21.9% hysterectomy, 44.4% lung surgery, 5.1% mastectomy, 3.5% prostate; only 1.8% of patients had reoperation for surgical arrest of bleeding without an accompanying diagnosis of haemorrhage and haematoma complicating a procedure (4 patients in the hysterectomy group, 4 patients in the lung surgery group); error bars represent 95% confidence intervals.

Table 1 Patient Characteristics*

	Patients without Bleeding Event N = 25,987		Patients with Bleeding Event N = 450		StdDiff
Patient demographics
Age category, N/%
18–24	333	1.3%	2	0.4%	9.8
25–45	3267	12.6%	45	10.0%	8.2
46–64	11,068	42.6%	191	42.4%	0.4
65 Plus	11,319	43.5%	212	47.1%	−7.0
Female, N/%	20,467	78.7%	352	78.2%	1.5
Geographic region, N/%
East of England	2436	9.4%	60	13.3%	−12.3
London	3368	13.0%	61	13.6%	−1.8
Midlands	3627	14.0%	51	11.3%	8.1
North	4521	17.4%	79	16.0%	−1.6
South Central	3837	14.8%	66	14.7%	0.3
South East Coast	3985	15.3%	58	12.9%	6.9
South West	3486	13.4%	58	12.9%	1.5
Yorkshire & The Humber	727	2.8%	17	3.8%	−5.6
Comorbidities, N/%
Anemia	747	2.9%	17	3.8%	−5.0
Cerebrovascular disease	306	1.2%	11	2.4%	−9.0
Chronic pulmonary disease	4031	15.5%	88	19.6%	−10.8
Chronic renal disease	888	3.4%	25	5.6%	−10.6
Congestive heart failure	299	1.2%	9	2.0%	−6.4
Connective tissue or rheumatic disease	528	2.0%	11	2.4%	−2.7
Dementia	138	0.5%	2	0.4%	1.5
Diabetes	3871	14.9%	73	16.2%	−3.6
Liver disease	290	1.1%	13	2.9%	−12.9
Myocardial infarction	222	0.9%	6	1.3%	−3.8
Peptic ulcer disease	120	0.5%	3	0.7%	−2.6
Peripheral vascular disease	393	1.5%	15	3.3%	−11.8
Medications, N/%
Anticoagulants	792	3.1%	27	6.0%	−13.9
Antiplatelets	3443	13.3%	94	20.9%	−20.3
Non-steroidal anti-inflammatory drugs	1895	7.3%	33	7.3%	0.0
Selective serotonin reuptake inhibitors	2772	10.7%	61	13.6%	−8.9
Selective norepinephrine reuptake inhibitors	324	1.3%	7	1.6%	−2.5
Measures of healthcare utilization and overall health status
Distinct prescription medications taken, mean/SD	3.2	3.0	3.9	3.5	−21.5
Face-to-face primary care consultations, mean/SD	9.4	8.0	10.7	9.7	−14.6
Prior hospitalization, N/%	3201	12.3%	59	13.1%	−2.4

Notes: *Patient demographics were measured as of the index admission; comorbidities, medications, and measures of healthcare utilization and overall health status were measured over the 12-month period prior to the index admission; SD, standard deviation; StdDiff, standardized difference (a value >|0.10| is indicative of imbalance between groups).

Table 2 Characteristics of Index Admission and Surgery*

	Patients without Bleeding Event; N = 25,987		Patients with Bleeding Event; N = 450		StdDiff
Index admission surgery type, N/%
Low anterior resection	2872	11.1%	85	18.9%	−22.0
Hysterectomy	5978	23.0%	114	25.3%	−5.4
Lung surgery	1511	5.8%	27	6.0%	−0.8
Mastectomy	12,611	48.5%	195	43.3%	10.4
Prostate surgery	3015	11.6%	29	6.4%	18.2
Index admission primary diagnosis type, N/%
Malignant neoplasm of the primary site	18,459	71.0%	336	74.7%	−8.3
Other malignant neoplasm	522	2.0%	10	2.2%	−1.4
Other neoplasms*	7006	26.9%	104	23.1%	9.0
Surgical approach, N/%
Open	21,856	84.1%	388	86.2%	−5.9
Minimally-invasive	4131	15.9%	62	13.8%	5.9
Index admission admitting status, N/%
Elective	25,552	98.3%	438	97.3%	6.8
Non-elective/emergency	435	1.7%	12	2.7%	−6.8
Index admission year, N/%
2010	5013	19.3%	73	16.2%	8.1
2011	4869	18.7%	104	23.1%	−10.8
2012	4791	18.4%	97	21.6%	−8.0
2013	4403	16.9%	74	16.4%	1.6
2014	3714	14.3%	53	11.8%	7.4
2015	2818	10.8%	45	10.0%	2.6
2016 (through February 29th)	379	1.5%	4	0.9%	5.5

Notes: *Other neoplasms included benign neoplasms, carcinoma in situ, and neoplasms of uncertain behavior; StdDiff, standardized difference (a value >|0.10| is indicative of imbalance between groups).

Figure 2 Adjusted association of bleeding events with post-procedural hospital length of stay*.

Note: *All difference in adjusted post-procedural hospital length of stay were statistically significant at P<0.05.

Abbreviations: CI, confidence interval; MID, mean incremental difference.

Figure 3 Adjusted association of bleeding events with post-procedural days spent in critical care*.

Note: *All difference in adjusted post-procedural days spent in critical care were statistically significant at P<0.05.

Abbreviations: CI, confidence interval; MID, mean incremental difference.

Figure 4 Adjusted association of bleeding events with in-hospital sequelae and in-hospital mortality during index admission*.

Notes: *Sequelae and mortality events were infrequent, which necessitated that we pool all patients from across the surgery groups when analyzing these outcomes; all difference in the incidence of sequelae and mortality were statistically significant at P<0.05.

Abbreviations: CI, confidence interval; MID, mean incremental difference.