Factors affecting quality of life after pelvic fracture

Abstract

Introduction

Functional outcome and health-related quality of life (HRQOL) after pelvis fracture is suboptimal; majority of the patients do not return to their preinjury activities. Many researchers reported that late morbidity in pelvis fracture is associated with severity of the fracture, associated trauma, pelvic fracture-related complications and methods of treatment.

Material and methods

One hundred and twelve patients with pelvis fracture who were treated either conservatively (n = 88) or surgically (n = 24) with a minimum of two years follow up were evaluated clinically and radiologically. The clinical outcome was evaluated using Majeed score and self-reported Short Musculoskeletal Function Assessment (SMFA). Their HRQOL was evaluated using the 36-item Short Form Survey (SF-36) and WHOQOL-BREF questionnaires. The fracture-displacement in the anterior or posterior pelvis ring was measured from the anteroposterior radiograph or inlet/outlet view.

Results

The average Majeed score was 76.65 ± 14.73 (range, 36 to 96). There were 81 patients with good to excellent outcomes and 31 patients with poor to fair outcomes. The average SF-36 Physical Component Summary (PCS) score was 47.71 ± 7.88 (range, 27.3 to 61.5) and SF-36 Mental Component Summary (MCS) was 49.20 ± 9.37 (range, 23.1 to 56.8). The functional level of the general population in the physical and mental domain was achieved in 48.23% and 65.3% of pelvic-fractured patients respectively. General population norms were achieved in 56.3%, 63.4%, 65.2% and 84.8% of patients in WHOQOL-BREF domain one, two, three and four respectively. The patients had significantly worse functional outcome and HRQOL if residual displacement was > 1 cm. Age, sex, associated injuries and injury mechanism were not affecting the HRQOL in patients with acceptable residual displacement of ≤ 1 cm.

Conclusion

Pelvic fracture with the residual displacement of ≤ 1 cm in the sacroiliac joint/symphysis pubis result in better functional outcome and HRQOL. Injury mechanism and associated injury have no impact on the HRQOL if the residual displacement is within the acceptable limit.

1. Introduction

Pelvic trauma in developing and underdeveloped nations is still predominated by high-velocity road traffic accidents and fall from height.¹ Most of the pelvis fractures are stable injuries; however, unstable and open pelvic fractures are associated with increased risk of mortality and morbidity.² Hemorrhage, infection, thromboembolism, soft tissue degloving injuries, associated genitourinary, gastroenterological and neurological injuries are immediate and early complications.^2,3 But late morbidity is associated with severity of the fracture, associated trauma, pelvic fracture-related complications and methods of treatment.^{4, 5, 6} The late sequel of pelvic fracture treatment can cause gait and postural disturbances, chronic pain, genitourinary dysfunction, persistent neurological deficits, and rectal dysfunction.^{7, 8, 9, 10, 11, 12, 13, 14, 15, 16}

Matta et al. reported chronic pain as the most common accompaniment of pelvic malunion and non-unions.¹⁷ McLaren et al. observed chronic pain in up to70% of patients with fracture displacement of more than 1 cm and only in 12% of patients with a displacement of less than 1 cm.¹⁸ Cranial displacement of hemipelvis creates a shortening of ipsilateral extremity with gait disturbances and sitting problems. The gait can also be affected due to the sciatic nerve or lumbar plexus injury resulting in foot drop or weakness of flexor muscles.^4,6 Sen and Lokesh in 2010 evaluated the outcome of pelvic ring fractures in 24 patients. They concluded that the involvement of sacroiliac joints in pelvic injuries affect the long term outcome adversely, more so if the residual displacement is more than 10mm¹⁹.

While evaluating functional outcome and health-related quality of life (HRQOL) in pelvis fracture, many studies reported that surgical fixation leads to a better return of activity.^{20, 21, 22} But few studies reported better HRQOL after conservative treatment.^23,24 Irrespective of the methods of treatment, the only predictor of favorable functional outcome is the anatomic reduction of the fracture/dislocation particularly in the sacroiliac joint.^18,19,25 The problems peculiar to these injuries in Indian scenario are late presentation due to lack of awareness and inadequate treatment facilities. Indians have a distinct cultural practice of sitting and squatting, where the actual functional capability can be more adversely affected and HRQOL may be significantly deranged due to residual changes in pelvic bone.^1,4,6 Accordingly, this study was designed to evaluate the functional outcome and HRQOL in pelvis fractured patients. The impacts of residual displacement on these outcome scores were also evaluated.

2. Material and methods

2.1. Patient recruitment

All patients with pelvic fractures managed in a tertiary care center between 2000 and 2010 were called and evaluated in the outpatient clinic to assess the functional outcome and HRQOL. The demographic profile, injury mechanism, fracture pattern, mode of treatment and postoperative complications were collected from the hospital records. For inclusion in the study, the subject must be ≥ 16 years of age and there should be complete clinical and radiological details with at least two-year follow up. Pathological fracture and open injuries were excluded for evaluation. Institutional ethics committee approval was obtained for the study and the consent of the patients were obtained before their recruitment. A total of 112 patients were recruited, of which 97 (86.6%) were male and 15 (13.4%) were female patients.

2.2. Patient evaluation

The patients were examined for pain, deformity and tenderness around the pelvic region. The range of motion of the hip and spine were measured. The gait pattern and muscle strength of hip abductor, adductor, flexor and extensors were recorded. The Majeed score was used to assess clinical outcome.²⁶ The patient-reported outcome evaluation was performed using the Short Musculoskeletal Function Assessment (SMFA) questionnaire.²⁷ The patients also answered questionnaires for the short-form (36) health survey (SF-36)²⁸ and WHOQOL-BREF questionnaires (WHO Quality of life-WHOQOL)²⁹ to evaluate their HRQOL.

The SMFA questionnaire is the shorter version of Musculoskeletal Function Assessment questionnaire with 46 items evaluation for dysfunction.²⁷ It has two sections: (i) 34 questions for the assessment of the patient’s function and (ii) 12 questions to assess how bothered patients are by their symptoms. The questionnaire can be used to evaluate the impact of a musculoskeletal condition on self-reported patient functioning and on the impact of the condition in impeding everyday life. This tool can measure the clinical impact of treatment in patients with musculoskeletal disease/injury, and provides a reliable and valid assessment of the health status of the individual patient. Higher total score represents greater degree of dysfunction or bother. SF-36 is a well-known, widely used and validated generic health outcome that consists of eight dimensions. Higher scores are associated with a better quality of life and state of health.²⁸ The WHOQOL-BREF score (26-item) is an abbreviated version of WHOQOL-100 questionnaires. It is a valid, reliable and responsive measure of HRQOL that evaluates an individual in four domains: physical, psychological, social and environment.²⁹ The WHOQOL-BREF Hindi version is a suitable instrument for comprehensive assessment of quality of life (QOL) in health care settings in Indian scenario.

Radiological evaluations were carried out using anteroposterior view, inlet view and outlet view of the pelvis. The displacement at sacroiliac joint (SI) and pubic symphysis were measured and recorded. The maximum value of the measurement among these three radiographic views was considered for evaluation. Anterior pelvic ring injuries included injury to the pubic symphysis and pubic rami whereas posterior ring injuries included posterior iliac wing injuries, sacroiliac fractures and sacroiliac displacements. Initial fracture/joint displacement (at the time of injury) and residual post-treatment fracture/joint displacement (at latest follow-up) were measured from anteroposterior view, inlet view and outlet view of the pelvis. In pelvic ring injury, there are ≥2 sites of injury. In APC1, APC2 and LC1 types of pelvis fracture, the displacement of the anterior pelvic ring was measured whereas, in APC3, LC2, LC3, VS and CM, the displacement of the posterior pelvic ring was measured. The displacement was categorized as mild (≤1 cm), moderate (>1–2 cm) and severe (>2 cm). Residual displacement of ≤1 cm (mild) was considered as the criteria of acceptable reduction. The average period of follow up was 3.76 years (2–17years).

2.3. Treatment

Out of 112 patients, 88 patients were managed conservatively (Fig. 1) and 24 patients were managed surgically. The decision to treat operatively or nonoperatively was based on the discretion of the senior surgeon (RKS). In the conservative management, pelvic binder and skeletal traction were applied for 6–12 weeks. However because of major associated trauma, comorbidities or excessive delay in presentation (>6 weeks), 22 patients could not be treated surgically (Fig. 2). A total of 24 patients were treated surgically (Iliosacral screws, symphyseal plating, iliac blade plating, external fixator, Fig. 3). The external fixator was applied in the emergency in 8 patients as a life-saving measure to control hemorrhage. The fixator was converted to internal fixation within fourteen days. However, because of pelvic-fracture related soft tissue and visceral injuries, four patients were managed with the external fixator as a definite fixation method.

Fig. 1.

Malunited lateral compression fracture of pelvis with moderate residual displacement following conservative treatment shows very good functional outcome in a 34-year male patient.

Fig. 2.

Conservatively treated vertical shear fracture with severe residual displacement causing limb length discrepancy and difficulty in seating in a 49-year old male patient.

Fig. 3.

Anteroposterior compression injury (APC III) of pelvis was reduced anatomically and stabilized with plates. This 24-year male patient had excellent functional outcome after 2 years follow up.

2.4. Statistical analysis

The effects of age, sex, associated injury, fracture-pattern/injury mechanism and residual displacement on functional outcome and HRQOL were analyzed using SPSS version 19.0. Descriptive statistics (mean, SD, skewness etc.) was used to describe the clinical characteristics, functional and radiological outcomes. Bi-variate association was evaluated using Pearson chi-square test. Multiple regression analysis was performed by modelling the outcome of interest (functional assessment) as a linear function of the predictor variables (fracture pattern, residual displacement), while adjusting for patient age, gender and length of follow up as potential confounding variables. The strength of association between residual sacroiliac (SI) displacement, pubic symphysis/rami displacement and functional outcome was carried out using Karl Pearson’s and Spearman’s rank correlation coefficient. Normal quintile plots were used to check the normality of data and appropriate transformations were used whenever required. Various comparisons were made either using t-test or analysis of variance (ANOVA).

3. Results

The average age was 32.58 ± 14.016 years (range, 16–65 years). The left-sided injury was observed in 43 patients and right-sided in 55 patients. In 14 patients, the injury was bilateral. The motor vehicle accident was the most common mode of injury; 105 (93.8%) patients had an injury from a motor vehicle accident while 7 (6.3%) patients had a history of fall from height. There were 24 anteroposterior compression (APC) injuries, 63 lateral compression (LC) injuries, 4 vertical shear (VC) injuries and 21 were complex injuries (Table 1). Associated injuries/fractures were observed in 84 patients (75%); of which, urological injury (n = 31, 27.7%) was most common. There were 11 patients (9.8%) with abdominal injuries, 19 patients (17%) with limb injuries, 9 patients (8%) with chest or head/maxillofacial injuries, 5 patients (4.5%) with neurological injuries and 7 patients (6.3%) with multisystem injuries. There were 41% (36/88) acceptable residual displacements in the conservative group whereas all patients in the surgical group had acceptable reduction/displacement (Fig. 3). The mean difference between the initial displacement and residual displacement in different types of pelvis fractures in both conservative and surgical treatment was significant except the vertical shear fracture which could not be reduced with the conservative treatment (Table 2).

Table 1.

Initial and residual displacements in pelvic-fractured patients after conservative treatment and surgical treatment.

Conservative treatment
Young and Burgess Classification	Mild dis. (≤1 cm)		Moderate dis. (>1–2 cm)		severe dis. (>2 cm)		Total
	Initial Displac.	Residual Displac.	Initial Displac.	Residual Displac.	Initial Displac.	Residual Displac.
APC1(Ant.Displac)	0	3	3	2	2	0	5
APC2 (Ant.Displac)	0	5	1	6	10	0	11
APC3(Post.Displac)	0	0	0	2	2	0	2
LC1(Post.Displac)	3	16	15	2	0	0	18
LC2(Post.Displac)	3	11	10	18	17	1	30
LC3(Post.Displac)	0	0	0	6	8	2	8
VS(Post.Displac)	0	0	1	2	2	1	3
CM(Post.Displac)	0	1	1	5	10	5	11
Total	6	36	31	43	51	9	88
Surgical treatment
APC3(Post. PelvicRing.Displac)	0	6	1	0	5	0	6
LC2(Post. PelvicRing.Displac)	0	5	0	0	4	0	5
LC3(Post. PelvicRing.Displac)	0	2	1	0	1	0	2
VS(Post. PelvicRing.Displac)	0	1	0	0	1	0	1
CM(PostPelvicRing.Displac)	0	10	1	0	9	0	10
Total	0	24	3	0	20	0	24

Table 2.

Mean-difference in initial and residual displacements in conservative and surgical group.

	N	Conservative Group		p-value1	N	Surgical Group		p-value2	p-value3 (conservative vs. surgical)
		Initial Displac. (cm)	Residual Displac. (cm)			Initial Displac. (cm)	Residual Displac. (cm)
APC I	5	2.14±0.19	1.14±0.32	0.042a	0	0	0
APC II	11	3.69±0.91	1.28±0.46	0.003a	0	0	0
APC III	2	2.7±0.14	1.8±0.00	0.018a	6	2.73±0.48	0.65±0.17	0.002a	0.071b
LC I	18	1.45±0.33	0.9±0.25	0.001a	0	0	0
LC II	30	2.2±0.73	1.34±0.56	0.001a	5	2.86±1.14	0.46±0.28	0.420a	0.098b
LC III	8	4.37±1.18	2.29±0.81	0.012a	2	2.5±0.707	0.5±0.00	0.080a	0.044b
VS	3	2.67±0.57	2.01±0.70	0.109a	1	4±0.0	0.7±0.0	0.005a	0.500b
CM	11	4.14±1.58	2.5±1.07	0.003a	10	4.39±1.41	0.7±0.22	0.005a	0.001b
TOTAL	88	2.70±1.30	1.499±0.779	0.010a	24	3.48±1.38	0.621±0.2245	0.001a	0.001b

^aWilcoxan signed rank test (p-value 1, p-value 2).

^bMann-Whitney test (p-value 3).

Outcome evaluation using Majeed score revealed an average score of 76.65 ± 14.73 (range, 36 to 96). There were 81 patients with good to excellent outcomes and 31 patients with poor to fair outcomes. Fifty-two of 81 patients with good to excellent outcomes had mild residual displacement (≤1 cm). Only 11 patients with severe displacement had excellent scores. The severity of displacement was shown to affect the Majeed score (p = 0.001, Table 3, Fig. 2). The effects of other variables such as the mechanism of injury and associated injury on the outcome in acceptable fracture reduction/residual displacement group were also evaluated. There was no association of injury mechanism (Young and Burgess Classification) to the outcome in the acceptable reduction group (p = 0.528). But, pelvic-fractured patients with associated injuries and acceptable reduction had lower Majeed score (p = 0.006). There were no associations of age and sex on clinical outcome (p > 0.05).

Table 3.

Effects of residual displacement severity on Majeed Score; effects injury mechanism and associate injuries on Majeed score in patients with acceptable residual displacement (≤1 cm).

		Majeed scores				Total	P-value
		Excellent (85–100)	Good (70–84)	Fair (55–69)	Poor (<55)
Residual Displacement	Mild	24	28	6	2	60	$\chi 2$ = 25.742, p-value 0.001∗
	Moderate	14	14	8	7	43
	Severe	11	0	4	4	9
	Total	39	42	18	13	112
Young and Burgess Classification (Mild displacement group)	APC1+LC1 Stable #	7	11	1	0	19	$\mathit{\chi }2$ = 5.222, p-value 0.528
	APC2+LC2 Part. Stable #	9	10	1	1	21
	APC3+LC3+VS + CM Comp. unstable #	8	7	4	1	20
	Total	24	28	6	2	60
Associate injuries (Mild displacement group)	Present	10 (25.0%)	22 (55.0%)	6 (15.0%)	2 (5.0%)	40 (100.0%)	$\chi 2$=12.53, p-value 0.006∗
	Absent	14 (70%)	6 (30%)	0 (0%)	0 (0%)	20 (100.0%)
	Total	24 (40.0%)	28 (46.7%)	6 (10.0%)	2 (3.3%)	60 (100.0%)

The average SF-36 Physical Component Summary (PCS) score was 47.71 ± 7.88 (range, 27.3 to 61.5) and SF-36 Mental Component Summary (MCS) was 49.20 ± 9.37 (range, 23.1 to 56.8). The functional level of the general population in the physical and mental domain was achieved in 48.23% and 65.3% of pelvic-fractured patients respectively. On comparing HRQOL of patients sustaining pelvic fracture with general population norm (student t-test), the scores were significantly low in pelvic-fractured patients (P-value<0.0001 for SF36 PCS and 0.034 for SF36 MCS). None of the patients could return to pre-injury functional level despite the good clinical outcome. The SF-36 PCS was significantly affected by residual displacement, but the SF-36 MCS was not affected. The injury mechanism and associated injury didn’t affect the SF-36 QOL scores in patients with acceptable residual displacement. The average SMFA score in this study was 31.59 ± 11.30 (range, 8 to 58.0) which is far below general population norms (9.26 ± 8.89, p-value 0.000). Only 2 patients were having SMFA scores comparable with their reference value (9.3). Posterior ring injuries had worse scores than anterior injuries. The average SMFA scores were 32.8 after conservative management and 25.59 after surgical management. Both, residual displacement severity and associated injuries were shown to affect the SMFA outcome (Table 4). However, there were no associations of age and sex on SF-36 and SMFA scores.

Table 4.

Short-Form 36 (SF-36) and Short Musculoskeletal Function Assessment (SMFA) scores in patients with different grades of residual displacement; Effects of injury mechanism and associated injuries on these HRQOL scores in patients with acceptable residual displacement (≤1 cm).

		SF 36- PCS		SF36-MCS		Total	P-value	SMFA		Total	p-value
		<47	≥47	<49	≥49			≤32	>32
Residual Displacement	Mild	19 (31%)	41 (68%)	18 (30%)	42 (70%)			48 (80%)	12 (20%)	60 (100%)	$\chi 2$=20.0, p-value = 0.0001∗
	Moderate	23 (53%)	20 (46%)	17 (40%)	26 (60%)			19 (44%)	24 (56%)	43 (100%)
	Severe	6 (66%)	3 (33%)	2 (22%)	7 (78%)			69 (61.0%)	43 (39.0%)	112 (100%)
	p-value	$\mathit{\chi }2$ = 7.136, p-value 0.028∗		$\chi 2$ = 1.547, p-value 0.461
Young and Burgess Classification (Mild displacement group)	APC1+LC1 Stable #	4 (21%)	15 (79%)	8 (42%)	11 (58%)	19 (100%)		16 (84.2%)	3 (15.8%)	19 (100.0%)	$\chi 2$=0.53, p-value = 0.765
	APC2+LC2 Part. Stable #	6 (28%)	15 (71%)	5 (23%)	16 (77%)	21 (100%)		17 (81.0%)	4 (19.0%)	21 (100.0%)
	APC3+LC3+VS + CM Comp. unstable #	9 (45%)	11 (55%)	5 (25%)	15 (75%)	20 (100%)		15 (75.0%)	5 (25.0%)	20 (100.0%)
	Total	19 (31%)	41 (69%)	18 (30%)	42 (70%)	60 (100%)		48 (80.0%)	12 (20.0%)	60 (100.0%)
	p-value	$\chi 2$ = 2.732, p-value 0.256		$\chi 2$ = 1.947, p-value = 0.378
Associated injuries (Mild displacement group)	Present	14 (25%)	35 (26%)	13 (32%)	27 (68%)	40 (100%)		29 (72.8%)	11 (27.2%)	40 (100.0%)	$\chi 2$=4.21, p-value 0.006
	Absent	5 (25%)	15 (75%)	5 (25%)	15 (75%)	20 (100%)		19 (95.0%)	1 (5%)	20 (100.0%)
	Total	19 (32%)	41 (68%)	18 (30%)	42 (70%)	60 (100%)
	p-value	$\mathit{\chi }2$=0.432, p-value 0.316		$\chi 2$ = 0.35, p-value 0.550

Data of all four domains of WHOQOL-BREF questionnaires showed non-normal distribution on one-sample Kolmogorov-Smirnov test (p-value 0.001). General population norms were achieved in 56.3%, 63.4%, 65.2% and 84.8% patients in domain one, two, three and four respectively, indicating a better outcome. The mean WHOQOL-BREF in domain1, domain 2, domain 3 and domain 4 scores were 66.57 ± 20.46 (vs. 71.1 ± 14.2 in general population norm), 60.04 ± 20.04 (vs. 63.0 ± 13.6 in general population norm), 70.54 ± 20.56 (vs. 68.8 ± 14.6 in general population norm) and 74.33 ± 16.74 (vs. 61.26 ± 12.8 in general population norm) respectively; so, the pelvic fractured patients were lacking in physical (domain 1) and psychological (domain 2) scores but not in social (domain 3) and environmental (domain 4) scores when compared to general population norms. There were no effects of age and sex on WHOQOL-BREF questionnaires. With increasing displacement, the WHOQOL-BREF score decreased significantly in physical (domain1), psychological (domain 2) and social relationship (domain 3) domains. But there was no significant association with environmental (domain 4) domain (Table 5). In the acceptable reduction group, fracture pattern does not affect any WHOQOL-BREF domain. Associated injury caused no significant reduction in any of the domain in the acceptable reduction group. There were excellent correlations among the disease-specific functional score (Majeed, SMFA) and HRQOL (SF-36 and WHOQOL-BREF) except the SF-36 MCS (Table 6).

Table 5.

The Heath-related Quality of Life assessment using WHOQOL-BREF Questionnaires (WHOQOL) in pelvic-fractured patients.

		WHOQOL1		WHOQOL2		WHOQOL3		WHOQOL4
		<67	≥67	<60	≥60	<70	≥70	<74	≥74
Residual Displacement	Mild	17 (28%)	43 (72)	15 (25)	45 (75	17 (29)	43 (71)	19 (31)	41 (69)
	Moderate	21 (48%)	22 (52)	20 (46)	23 (54	20 (46)	23 (54)	22 (51)	21 (49
	Severe	7 (77%)	2 (23)	6 (66	3 (34	5 (55)	4 (45)	1 (11)	8 (89)
	Total	45 (40)	67 (60)	41 (36	71 (64	42 (37)	70 (63)	42 (37)	70 (63)
	p-value	$\mathit{\chi }2$ = 10.13, p = 0.006∗		$\mathit{\chi }2$ = 5.780, p = 0.049∗		$\mathit{\chi }2$ = 4.89, p = 0.087		$\mathit{\chi }2$ = 6.97, p = 0.049∗
Young and Burgess Classification (Mild displacement group)	APC1+LC1 Stable #	6 (31%)	13 (69%)	9 (47%)	10 (53%)	7 (36%)	12 (63%)	10 (52%)	9 (48%)
	APC2+LC2 Part. Stable #	5 (23%)	16 (77%)	2 (9%)	19 (91%)	4 (19%)	17 (81%)	4 (19%)	17 (81%)
	APC3+LC3+VS + CM Comp. unstable #	6 (30%)	14 (70%)	4 (20%)	16 (80%)	6 (30%)	14 (70%)	5 (25%)	15 (75%)
	Total	17 (28%)	43 (72%)	15 (25%)	45 (75%)	17 (28)	43 (71%)	19 (31%)	41 (69%)
	p-value	$\mathit{\chi }2$=0.338, p = 0.845		$\mathit{\chi }2$=4.966, p = 0.083		$\mathit{\chi }2$=1.59, p = 0.450		$\mathit{\chi }2$=5.81, p = 0.055
Associated injuries (Mild displacement group)	Present	12 (30%)	28 (70%)	10 (25%)	30 (75%)	12 (30%)	28 (70%)	14 (355)	26 (65%)
	Absent	5 (25%)	15 (75%)	5 (25%)	15 (75%)	5 (25%)	15 (75%)	5 (25%)	15 (75%)
	Total	17 (28%)	43 (72%)	15 (25%)	45 (75%)	17 (28%)	43 (71%)	19 (31%)	41 (68%)
	p-value	$\mathit{\chi }2$=0.164, p = 0.685		$\mathit{\chi }2$=0.260, p = 0.564		$\mathit{\chi }2$=0.164, p = 0.685		$\mathit{\chi }2$=0.616, p = 0.432

Table 6.

Correlations between disease specific Functional score (Majeed, SMFA) and Health related Quality of life scores (SF-36, WHOQOL-BREF questionnaires-WHOQOL).

Pearson correlation		Majeed score	SMFA	SF-36 PCS	SF-36 MCS	WHOQOL1	WHOQOL2	WHOQOL3	WHOQOL4
Majeed Score	r	1	−0.678	0.465	0.162	0.564	0.444	0.391	0.093
	p-value		0.001∗	0.001∗	0.130	0.001∗	0.001∗	0.001∗	0.388
SMFA	r	−0.678	1	−0.281	−0.018	−0.347	−0.364	−0.238	0.053
	p-value	0.001∗		0.008∗	0.867	0.001∗	0.001∗	0.025∗	0.625
SF36 PCS	r	0.465	0.281	1	0.012	0.314	0.147	0.234	0.091
	p-value	0.001∗	0.008∗		0.913	0.003∗	0.172	0.028∗	0.398
SF36 MCS	r	0.162	−0.018	−0.012	1	0.216	0.335	0.109	0.164
	p-value	0.130	0.867	0.913		0.043∗	0.001∗	0.310	0.127

r: correlation coefficient.

A total of 65 patients (58.03%) developed complications. The details of the complications were as follows: urological complications (n = 26, 23.2%), persistent pain (n = 12, 10.7%), impotence (n = 9, 8.0%), limp (n = 5, 4.5%), deep venous thrombosis (n = 5, 4.5%)), neurological complications (n = 5, 4.5%) and infection (n = 3, 2.7%). Two patients had superficial infections and needed prolonged course of antibiotics. One patient with deep infection was treated with debridement and antibiotics.

4. Discussion

The main findings of this study were that patients with a residual displacement of ≤1 cm in the sacroiliac joint and symphysis pubis had a better functional outcome and health-related quality of life. Surgical treatment leads to better reductions in pelvis fracture. The physical and psychological HRQOL in these pelvic-fractured patients were inferior to the general population norm. Associated injury and injury pattern/mechanism did not affect the HRQOL if the residual displacement was ≤1 cm.

Because of the high-velocity trauma and associated visceral and soft tissue injuries, pelvis fracture patients remain disabled for longer time.¹ Besides, pelvis, being the junction of the axial and appendicular skeleton, impairs the mobility of the patient significantly. Hence, pelvis fractures have a substantial impact on physical, psychological, social and occupational health along with financial burden.^4,6,30,31 Most of the studies reported worse functional and subjective outcomes after pelvic ring fractures when compared to the population norms over a long run.^{12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,30,31} Suzuki et al. reported a worse SF-36 score in unstable pelvic ring injuries after two to nine years.¹⁵ They reported average Majeed score of 79.7 and the average PCS of the SF-36 was 13.4 points worse than that of the general population norm. Holstein et al. reported lowered EQ (5D) scores among pelvic-fractured patients compared to the general population norms.²⁴ Oliver et al. reported 14% impairment in physical outcome and 5.5% impairment in mental outcome score in patients of pelvis fracture when compared to general US population.²¹ Banierink et al. also observed substantially lower physical functioning (mean SMFA function score 22 vs. 12) and HRQOL (mean EQ-5D 0.76 vs. 0.87) after 4 years of pelvis fracture when compared to their peers from the general population.³⁰ In agreement with the literature, we observed lower functional score (Majeed score 77) and HRQOL (PCS 48, MCS 49) in our patients. One important observation of this study was that SF-36 MCS was low compared to general population norm but it was not affected by the severity of displacement or associated injuries.

The treatment methods, operative or nonoperative, have not shown any association with the outcome.^{13, 30} Miranda et al. reported 81% of patients returning to their work irrespective of operative or non-operative management in a cohort of 78 patients with pelvis fracture.¹³ Lykomitros et al., in 2009 retrospectively evaluated 16 patients with traumatic sacral fractures for 2 years using SF-36 questionnaires filled up by the patients. Those who were treated conservatively achieved the best scores in every domain of the SF-36 questionnaires than operative patients.²³ A recent study by Banierink et al. also didn’t find any difference in physical functioning and HRQOL between operatively and non-operatively treated patients.³⁰ Seventy-eight percent of patients in their series were managed nonoperatively.³⁰ Because of the associated injuries, comorbidities and excessive delay in referral, we adopted conservative treatment in 22 patients who could have been benefitted by surgical intervention. Nevertheless, the outcome of the patients in this series is comparable to that of literature. With the conservative treatment, 20 patients with moderate displacement could be converted to acceptable reduction, whereas, 42 patients were converted from severe displacement to moderate displacement. This could be possible because of a strict protocol of prolonged pelvic traction and pelvic binder application with regular radiographic monitoring. These patients remained in the hospital for six weeks. In the operative group, 24 patients from moderate or severe displacement group could be converted to a mild displacement group. There are several reports where surgical outcome have been associated with poor HRQOL. Probably, more sever and displaced fracture were selected for the operative intervention and hence the poor outcome. Holstein et al. reported a significantly reduced HRQOL in patients who were managed surgically, and several other articles had also similar conclusion.²⁴

There are controversial reports on the factors predicting the functional outcome and HRQOL in pelvis fracture (Table 7).^{12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,30,31, 32, 33, 34} While many studies reported age and sex have no association on the outcome,^{15, 16, 17,26} few authors reported significant association.^24,32 Holstein et al., in 2013 reported that age is an independent risk factor of HRQOL.²⁴ However, their conclusion should be interpreted cautiously as they mention specifically about geriatric pelvic fracture (0.78 versus 0.7–0.8 in elderly and 0.8–0.9 in younger individuals) where there is increasing need of community support and institutionalization in the postoperative period along with the increased risk of intraoperative bleeding and perioperative mortality. The average age in our series was 33 years and none of the patients was more than 65 years of age. Many studies didn’t report the fracture pattern as a risk factor for poor outcome.^15,21,24 The fracture severity was assessed using the Tile’s classification in these reports. To the best of our knowledge, Young and Burgess Classification have been used in limited studies.^26,35 The Young and Burgess classification is a modification of the Tile classification.^36,37 It is the recommended and most widely used classification system as it considers the force type, severity, and direction, as well as instability in pelvis fracture. The fracture in APC I and LC I is equivalent to Tile group A, APC II and LC II is equivalent to Tile group B and APC III, LC III, VS and CM is analogous to Tile group C. Accordingly, This study didn’t observe any association between fracture pattern/injury mechanism and the functional outcome/HRQOL in patients with the residual displacement of ≤ 1 cm.

Table 7.

Studies on health related quality of life in pelvic-fractured patients.

Authors	No of patients	Age (years)/sex (mean/median)	Treatment	Outcome scores used	Follow-up (mean/range in months)	Conclusions
Ayvaz et al.33	20	32 (11–66)/11:9	CR and percutaneous fixation of unstable pelvis fracture	Majeed, SF-36, Iowa pelvic score, pelvic outcome score	24–48	SF-36 score comparable with that of the population norm.
Suzuki et al.15	57	42.2/28:29	23 treated conservatively, 22 with external fixation, 12 with internal fixation (Tile B and C)	SF-36, Iowa pelvic score, Majeed scores	24–107	SF-36 score worse than normal population norm, Long-term functional outcome was not associated with ISS, fracture type and location. There is a close correlation between neurologic injury and functional outcome.
Holstein et al.24	172	8–88 years (median, 47/103: 69	125 (73%) treated surgically in both stable and unstable pelvis fracture	EQ-5D	12–72	Lower ED-5D score compared to population norms. Higher age, complex trauma, and surgery had a reduced HRQOL.
Miranda et al.13	80	38 (17–68)/50:30	48 (61%) were treated with external fixation, and 32 (39%) were treated nonoperatively.	SF-36, General Health Survey, the Iowa pelvic scores, and others	60	Approximately 75% of patients returned to their previous sexual function and 80% to their previous occupation. Results of nonoperative, open reduction and internal fixation, and external fixation to treat pelvic ring injuries are similar.
Bosch et al.34	37	34.7 (15–66)/36:11	7 underwent ORIF of pubic arch, 10 were treated with a combination of anterior ORIF with additional external fixation, 19 underwent internal fixation of both anterior and posterior arch, 11 patients underwent percutaneous posterior screw fixation combined with anterior external fixation (Tile B and C)	Majeed, SF-36	35.6	Limitations in functioning are reported, even after long-term follow-up. Patients treated with combined internal fixation anterior as well as posterior scored a better outcome compared with combined internal and external fixation
Oliver et al.21	35	33.5/21:14	29 ant and post fixation; 31 ant fixation and 33 post fixation (Tile B and C)	SF-36	24	14% impairment in PCS and 5.5% impairment in MCS compared with the normal population.
Mullis et al.25	23	–	Symphyseal fixation and iliosacral screw fixation in all vertical shear fractures	Majeed, SF-36, Iowa pelvic score, SMFA	12	Treatment of vertically displaced, pure SI joint dislocations, an anatomic reduction (whether closed or open), followed by ISS fixation should be the goal
Borg et al.8	54	34/28:26	ORIF in 20, percutaneous in 14 and combined in 20 patients.	SF-36, LiSat 11	24	Patients reported substantially lower HRQOL for both physical andmental domains, when compared with a reference population after surgical treatment of pelvic ring fractures.
Banierink et al.30	192	54/108: 84	150 nonoperative; 42 operative	SMFA, EQ-5D	52	Long-term physical functioning and HRQOL in patients with pelvis fracture is fair, but significantly decreased in comparison with their peers from the general population.
Present study	112	33/97:15	88 nonoperative, 24 operative	Majeed,SF-36, SMFA	24	Residual displacement of >10 mm is associated with poor outcome and HRQOL. Associated injury and injury mechanism have no impacts on the HRQOL if the residual displacement is within the acceptable limit.

Residual posterior pelvis injury displacement, associated injury and neurological displacement have been associated with poor outcome.^{4,6,12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,30,31} The importance of posterior ring injuries as a major factor affecting the outcome is quite explicit. In Tile’s series, 40% of the patients had residual pain and among these 90% were having posterior lesions.³⁶ Dujardin et al. have shown that posterior ring injuries had worse outcome.¹² In the present study, the posterior ring injuries had a poorer outcome compared to anterior pelvic ring injuries. This difference can be attributed to the crucial role of the sacroiliac joint in weight transmission, unlike anterior injuries. In this series, pain was present in 80% of patients in posterior ring injury and only in 20% of the anterior pelvic ring injury. The cause of pain in posterior pelvic injury could be because of sacroiliac degeneration even after adequate reduction. The limp was observed in 40% of patients of posterior ring injury but no limp was observed in the anterior pelvic ring injury. It has been mentioned that methods of treatment have no impact on the outcome among sacroiliac disruption as long as fracture reduction is achieved. Henderson and McLaren et al. reported that displacement of >1 cm in the posterior pelvis was associated with increased long-term pain and disability.^18,38 Despite an acceptable residual displacement, few studies failed to find a satisfactory outcome.^14,16 These studies reported the importance of associated visceral and soft tissue injuries in pelvis fractures.³⁹ Associated injuries, particularly those are pelvic fracture-related, have a significant association with the poor outcome even with the excellent radiological outcome. The study by Pohlman et al.¹⁶ reported excellent reduction with less than 5 mm residual posterior displacement in 80% of patients with Tile type C injury; but, only 27% patients reported good or excellent outcome. Another study reported that the outcome of unstable pelvic fracture may be inferior to stable fracture not only because of instability and asymmetry but could be due to severity and amount of damage to the soft tissues.⁴⁰ Korovessis et al.³⁹ also reported that the functional outcomes were superior in patients without associated injuries in pelvis fracture. With the assumption that patient with mild displacement on radiographs were the ones who had the possibility of best outcomes, rest of the analysis for evaluating the effect of associated injuries was conducted on these mild displacement group patients only. In this study, there were 47% associated pelvic injuries, of which 27.7% were urologic injuries. These associated visceral or neurological injuries usually do not recover completely even in the long run and probably affect the HRQOL. However, disease-specific outcome scores may ignore these issues and overrate functional outcome.⁴¹ Probably HRQOL assessment in pelvis fracture has more importance than the functional score for overall general health measurement. Consequently, this study used two measuring tools (SF-36 and WHOQOL-BREF questionnaires) for HRQOL assessment. Although neurological injuries have also been reported as an independent risk factor poor HRQOL,¹⁵ the incidence was quite low in our study. There were no associations of associated injuries and injury mechanism/fracture pattern on these HRQOL measures, particularly when the residual displacement was ≤ 1 cm. Patients with mild residual displacement had a probably lesser degree of initial injury and hence less severe associated injuries. Perhaps the only significant predictor of HRQOL in pelvic fracture patients is the amount of residual displacement.

There are several limitations in this study. First of all, the retrospective data collection from the medical records could have caused missing of the data. The injury severity score was not assessed and only the associated injuries were mentioned. Secondly, many patients were managed conservatively because of late referral and pelvic-fracture related soft tissue or visceral complications. Thirdly, the overall treatment of pelvis fracture in this study was based on the management protocol prevailed at that particular time. Despite these limitations, the functional outcome evaluation and HRQOL assessment in this study were correlated to residual displacement when the fracture has already healed or dislocation has been stabilized. We believe that sequelae of associated injuries and residual displacement affect the functional outcome in the long run. But the residual displacement of >10 mm is the only predictor of poor outcome in HRQOL measures.

To conclude, patients with the residual displacement of ≤1 cm in the sacroiliac joint/symphysis pubis result in better functional outcome and HRQOL. This desired reduction can be better achieved by surgical treatment. Despite an acceptable radiological reduction, many patients cannot return to their preinjury activity level. Residual displacement in the sacroiliac joint or symphysis pubis and associated injuries significantly predict the functional outcome. Injury mechanism and associated injury have no impact on the HRQOL if the residual displacement is within the acceptable limit.

Source of support

None.

Ethical consideration of this manuscript

This study was conducted in PGIMER Chandigarh by Dr. VV under guidance of Dr. RKS, Dr. SA and SS as a thesis for M.S. degree fulfilment. The study was approved by the Institute Ethics Committee, PGIMER Chandigarh. Written consents were obtained from all the patients before their recruitment in the study.

Declaration of competing interest

None.