Acetabulum fractures in elderly: A systematic review of fracture pattern and treatment

Abstract

Purpose

Acetabulum fractures are being increasingly seen with low impact injuries in elderly patients. This article aimed to study systematically literature on geriatric acetabulum fractures. Objectives of this systematic review were to study (1) demography of patients, common mechanisms of injury and types of fracture patterns commonly seen in elderly patients, (2) treatment used for these fractures in literature and (3) mortality rates in elderly with these fractures.

Methods

Systematic search was carried out in May 2020 using predefined search strategy for all studies published in the English language in last 20 years. Literature search and data abstraction was done by two independent reviewers.

Results

After screening of all abstracts, a total of 48 studies were included in the systematic review. In total there were 7876 geriatric patients with acetabulum fractures. Mean age of the patients was 72.47 years. There were 4841 males (61.5%). Fall from low heights was the most common mechanism of injury, present in 47.12% patients followed by motor vehicular accidents in 28.73%. Most common fracture pattern was both column fracture, seen in 19.03% patients, followed by anterior column and posterior hemitransverse fracture in 17.23%, anterior column fractures in 17.13%, and posterior wall fractures in 13.46% patients. Out of total 5160 patients for whom details of treatment were available, 2199 (42.62%) were given non-operative treatment, 2285 (44.28%) were treated with ORIF of acetabulum fracture, 161 (3.12%) were treated with percutaneous fixation and 515 (9.98%) were treated with primary THA. Gull sign, femoral head injury and posterior wall comminution were associated with poorer prognosis after ORIF and may form an indication for a primary THA.

Conclusion

Literature on treatment of geriatric acetabulum fractures is not enough to draw any definite conclusions. There is limited evidence from current literature that surgery could be considered a safe treatment option for displaced acetabulum fractures in elderly. Primary THA can provide early mobility and reduce chances of resurgery in fracture patterns where restoration of joint surface may not be possible.

1. Introduction

Acetabulum fractures were initially discussed in context to high energy trauma in adults, but are now becoming increasingly common with low impact injuries in elderly.^1,2 They behave differently in elderly as compared to younger patients.^1,3 The number of elderly patients is rising across the globe. There has been a consequent increase in number of geriatric acetabular fractures. Geriatric acetabulum fractures have received lesser attention than other fractures in elderly, such as hip fractures.

Three most common treatment modalities for these fractures are nonoperative treatment, open reduction and internal fixation (ORIF) and primary total hip arthroplasty (THA).^3,4 Treatment of elderly is complicated because of associated co-morbidities and poor bone quality.^4,5 Elderly patients tolerate trauma from injury or from surgery poorly as compared to the younger patients. There are no definite guidelines on management of these injuries. Indications of surgical treatment and the type of surgical treatment needed are not clearly defined.

Previous reviews on acetabulum fractures in elderly were not systematic and also included studies not restricted to elderly population.^4,5 This article aimed to study systematically literature on geriatric acetabulum fractures. Objectives of this systematic review were to study (1) demography of patients, common mechanisms of injury and types of fracture patterns commonly seen in elderly patients using Judets and Letournel classification, (2) treatment used for these fractures in literature and (3) mortality rates.

2. Material and methods

Systematic search was carried out in confirmation with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Literature search was carried out in May 2020 by two different reviewers using four different online databases: PubMed (http://www.ncbi.nlm.nih.gov/pubmed), EMBASE (http://www.elsevier.com/online-tools/embase), Google Scholar, and the Cochrane database (http://www/.cochrane.org), for all studies published in the English language in past 20 years. Search was carried out using terms: geriatric, elderly, low energy, fragility fracture, osteoporosis, acetabulum and acetabular fractures. These keywords were combined for search using appropriate boolean operators. References of all the included studies were also screened. ‘Similar articles’ and ‘cited by’ options on PubMed were also used. Reference lists of review articles were also studied for potentially relevant papers.

We included studies whose study-population was limited to elderly patients, defined as more than 55 years of age (based on a preliminary review of literature on age cut-offs used in different studies), or studies from which information on elderly population could be extracted separately. We excluded studies on pelvic fractures, case reports, conference abstracts, case series with less than 5 cases, review articles and articles on surgical techniques. There were no randomized studies available on this topic. A broad variety of studies were included, like prospective and retrospective studies, observational and experimental studies. If there were more than one studies by the same author or the same center, the latest study was taken for pooling the data.6, 7, 8, 9, 10, 11, 12, 13

2.1. Data abstraction

Data was extracted from the included studies by 2 independent reviewers using a pre-determined form. Information was abstracted on age, gender, fracture classification, treatment, outcomes and complications. Continuous variables were extracted as means and standard deviations wherever available.

3. Results

3.1. Studies and samples

The total number of abstracts screened initially was 2352. Fig. 1 shows the PRISMA flow-diagram for literature search. After screening them, a total of 48 studies were included in the systematic review (Table 1). In total, there were 7876 geriatric patients with acetabulum fractures. Mean age of the patients was 72.47 years. There were 4841 males (61.5%). Details of the studies and the patients are summarized in Table 1.

Fig. 1.

The PRISMA flow-diagram for literature search in the systematic review.

Table 1.

Details of studies and patients included in the systematic review.

Studies	Study objective	Cut-off for age in years	Displaced/Undisplaced fractures	Study type	Number of patients	Mean age in years±standard deviation (range)	Number of males (percentage)	Duration of study	Follow-up in months (range)
Anglen et al. 200319	Outcomes of ORIF	60	Displaced fractures	Retrospective, non-comparative	48	71.6 (61–88)	42	July 1992–August 1999	37 (1–114)
Archdeacon et al. 201324	Outcomes of ORIF	70	na	Retrospective, non-comparative	39	80 (70–96)	24	November 2000 to December 2009	34 (12–127)
Baker et al. 201916	Outcomes of conservatively treated fractures	65	25% minimally displaced, remaining displaced	Retrospective, non-comparative	49	80 (65–94)	29	June 2008 and June 2016	12
Beaule et al. 200446	Outcome of primary THA	50	Displaced	Retrospective, non-comparative	10	61 (50–85)	5	Over 6 years	36 (24–55)
Bible et al. 201420	1-year mortality of operative and conservative treatment	60	na	Retrospective, comparative	86	71.1 ± 7.1	65	1998–2009	16 (0–112)
Boelch et al. 201640	Outcomes of ORIF vs primary THR	>54	na	Retrospective, comparative	32	75.2 (59–92)	22	2005–2015	4.5 (2.5–9.5)
Boraiah et al. 200925	Primary THA	55	Displaced fractures	Retrospective, non-comparative	18	72 (55–86)	10	1997–2007	46.8 (12–121.2)
Borg et al. 201942	Outcome of primary THA	na	Displaced fractures	Retrospective	13	76.5 (64–89)	8	2003–2014	24
Boudissa et al. 201933	Outcomes of ORIF vs primary THR	60	Displaced fractures	retrospective, comparative	82	73 (63–101)	62	2005–2014	24
Carroll et al. 201021	Outcomes of ORIF	55	na	retrospective, non comparative	93	67 (56–89)	51	1992–2005	60 (24–188)
Chakravarty et al. 201414	Outcomes of percutaneous column fixation & THA	na	na	retrospective, non comparative	19	77 (57–90)	13	2005–2011	22 (2–80)
Deren et al. 201717	Sarcopenia in geriatric acetabulum fractures	60	na	Retrospective	99	74.31	61	2005–2014	16–120
Enoscon et al. 201458	Primary THA	>60	displaced	Prospective cohort	15	75.5 (63–84)	8	2001–2008	48
Ernstberger et al. 202030	Comparison of open, percutaneous and non-operative treatment from German Pelvic Injury Register	60	Minimally displaced	Retrospective	608	77.3 ± 9.7 (60–100)	381	2008–2018	na
Ferguson et al. 20101	Radiological characteristics	60	displaced	Retrospective	235	70 (60–98)	160	1980–2007	na
Firoozabadi et al. 201631	Descriptive study on geriatric acetabulum fractures	65	na	Retrospective comparative	156	77.5 years (65–97)	113	5 yrs	na
Gary et al. 20117	Outcomes of Percutaneous fixation	60	displaced	Retrospective, non comparative	79	73±1yr	52	1994–2007	46.8
Giunta et al. 201818	Outcome of primary THA	>60	na	Retrospective, non comparative	27	68.5 ± 8.1 (60–84)	23	2010–2015	48 (12–84)
Guerado et al. 201247	Outcome of primary THA	80	na	Case series	4	≥80	na	na	12
Herath et al. 20192	Outcomes of operatively treated vs non operatively treated fractures from German Pelvic Injury Register	60	na	Retrospective	1914	76.6 ± 9.5	1193	2002–2017	na
Herscovici et al. 201047	Outcomes of primary THA	na	na	Retrospective, non comparative	22	75.3 (60–95)	12	1995–2005	29.4 (13–67 m)
Jeffcoat et al. 201248	Outcomes of ORIF, comparison between two approaches	55	na	Retrospective	41	67 (56–85)	31	1992–2006	63 (24–188)
Kim et al. 201549	Low vs high energy injury	60	na	Retrospective	186	70.2 (60–96)	129	2001–2012	na
Kim et al. 202015	Outcomes of ORIF	65	na	Retrospective	2471	76 ± 7	1436	2002–2014	na
Laflamme et al. 201538	Outcomes of ORIF, Gull sign	ns	Displaced	case series	9	64.3 (50–84)	5	na	33.6 (12–74).
Li et al. 201450	Outcomes of ORIF vs primary THR	60	Displaced	Retrospective	52	69.9 (60–90)	43	2000–2008	72
Lin et al. 201551	Outcomes of ORIF vs primary THR	60	Displaced	Retrospective	33	66 (47–92)	15	1996–2011	67.2 (1–171.6)
Lont et al. 201952	Outcomes of ORIF vs primary THR	55	na	Retrospective comparative	59	70 (56–92)	42	2008–2017	ORIF 50.4 (0–108); THA 16.8 (0–72)
Malhotra et al. 201353	Outcome of primary THA	55	Displaced	Case series	15	64.5 (57–69)	13	2000–2005	81.5 (62–122)
Manson et al. 201759	Outcomes of operatively treated vs non operatively treated fractures	60	Displaced	Retrospective observational study.	269	more than 60	200	January–December 2009	na
Mears et al. 20028	Outcome of primary THA	na	Displaced	Retrospective	57	69 (26–89)	30	1985–1997	97.2 (24–144)
Miller et al. 201060	Radiological outcome	55	na	Retrospective non comparative	45	67 (59–82)	35	92–2006	72.4 (24–188)
Mitchell et al. 201822	Sarcopenia as the predictor of 1-year mortality	60	na	Retrospective	146	70.1 ± 7.4	107	2003–2014	na
Moushine et al. 200411	Outcome of primary THA	na	Displaced	Case series	18	76 (65–93)	na	1998–2001	36 (12–46)
Ortega-Briones et al. 201763	Outcome of primary THA	na	na	Retrospective, non comparative	24	77.4 yr (62–92)	9	na	na
Otoole et al. 201461	Outcomes of ORIF	60	Displaced	Retrospective	61	69 (60–88)	na	2001–2006	52.8 (13–96)
Papadakos et al. 201427	Descriptive study on low energy fractures	na	na	Retrospective	71	67 ± 19.1	50	2005 to 2008	na
Resch et al. 201745	Outcome of primary THA	65	Displaced	Retrospective	30	79.9 (65–92)	15	2009 to 2014	6
Rickman et al. 201454	elderly osteoporotic fracture with fracture fixation and THA	60	na	Case series	24	77 (63–90)	16	2009–2012	24 (8–38)
Rodrı´guez et al. 201256	Outcome of primary THA	65	na	Case series	6	77 (70–85)	3	2008–2011	24
Ryan et al. 201734	conservative	60	Displaced	Case series	27	76 ± 8.7 (60–94)	na	2002–2012	2.2y (1-6.5y)
Salama et al. 201626	Outcome of primary THA	>35	na	retrospective, non comparative	18	66.1 (35–81yr)	12	2011–2014	21.7 m (12–36)
Schnaser et al. 201632	Rate of conversion of acetabular fracture to THA	>60	na	Retrospective	171	72.5 yr (60–94)	124	2001–2011	13.4
Tidermark et al. 200357	clinical & functional outcome in e; der; y	>55	displaced	Retrospective	10	73 (57–87)	7	1993–1999	38 m (11–84)
Walley et al. 201735	elderly and severely comorbid patient	65	na	Retrospective	86	79.7 (65–94)	na	2005–2014	15.14 (1–60)
Weaver et al. 201844	Outcomes of ORIF vs primary THR	65	na	Retrospective, comparative	70	76	38	2002 to 2009	22 months (range 6–89 months)
Wollmerstädt et al. 202064	Outcomes of operatively treated vs non operatively treated fractures	60	na	Retrospective	176	78 ± 10	103	2008 to 2016	68 months, SD 26, range, 24 to 129
Zha et al. 201713	Effect of age on outcomes of ORIF	60	displaced	Retrospective	53	72.85 (60–90)	44	2004 to 2011	52.5 ± 24.1
	na not available, CHP combined hip procedure, ORIF open reduction internal fixation, THA total hip arthroplasty
Total					7876	72.47	4841 (61.5%)

3.2. Mechanism of injury

Mechanism of injury was presented in 29 studies (Table 2). Overall, fall from low heights was the most common mechanism of injury, present in 810/1719 (47.12%) patients, followed by motor vehicular accidents in 494/1719 (28.73%) patients.

Table 2.

Details of mechanism of injury in the included studies.

	Total	Fall from standing height	Fall from greater height	Bike/Motorcycle	Automobile/motor vehicular	Pedestrian	Others
Anglen et al. 200319	48	11	7	0	24	0	6
Archdeacon et al. 201324	39	32	3	0	3	0	1
Beaule et al. 200446	10	7	0	0	0	0	3
Bible et al. 201420	86	18	23	2	32	3	8
Borg et al. 201942	27	15	1	0	10	1	0
Boudissa et al. 201933	82	74	0	0	0	0	8
Carroll et al. 201021	93	47	0	0	31	13	2
Chakravarty et al. 201414	19	14	0	0	5	0	0
Ferguson et al. 20101	235	117	0	12	88	15	3
Firoozabadi et al. 201631	156	48	62	4	36	2	4
Herscovici et al. 201047	22	17	0	0	5	0	0
Jeffcoat et al. 201248	41	32	0	0	9	0	0
Kim et al. 201549	186	56	36	12	54	22	6
Li et al. 201450	52	15	19	0	0	17	1
Lin et al. 201551	33	11	1	0	16	5	0
Lont et al. 201952	59	45	3	5	6	0	0
Malhotra et al. 201353	15	9	0	0	6	0	0
Mitchell et al. 201822	146	21	27	5	78	7	8
Moushine et al. 200411	18	10	0	0	8	0	0
Otoole et al. 201423	61	18	10	0	33	0	0
Papadakos et al. 201427	71	60	0	0	0	0	11
Resch et al. 201745	30	26	3	0	1	0	0
Rickman et al. 201454	24	22	1	0	1	0	0
Rodrı´guez et al. 201256	6	5	0	0	0	0	1
Ryan et al. 201734	27	21	2	0	0	0	4
Tidermark et al. 200357	10	10	0	0	0	0	0
Weaver et al 201844	70	40	0	0	29	0	1
Zha et al. 201713	53	9	7	7	19	11
Total	1719	810 (47.12%)	205 (11.92%)	47 (2.73%)	494 (28.73%)	96 (5.58%)	67 (3.89%)

Elderly patients are more likely to sustain injury from low injury mechanisms. Chakravarty et al.¹⁴ had average age of 80 years for the low-energy group, and 65 years for the high-energy group. In a series of 1309 patients, Fergusson et al.¹ found that about fifty percent of all injuries in elderly were caused by fall from standing height. In patients less than 60 years of age, falls accounted for only 17.7% of the injuries. Motor vehicular accidents accounted for about 66% injuries in younger patients, and only 37.4% in patients over 60 years of age. Kim et al.¹⁵ studied 183 acetabulum fractures in patients over 60 years of age and found fall from height of less than 1 m in 30.1%, fall from greater height in 19.4% and motor vehicular accidents in 29% patients. Baker et al.¹⁶ had 45 out of 49 patients who suffered fall from standing height. In a few series, high energy mechanisms were found to be more common. Deren et al.¹⁷ found that 67 out of 99 patients had fractures with high-energy mechanisms of injury. Giunta et al.,¹⁸ in their series, found ten fractures were due to low-energy mechanisms, and 17 fractures were due to high-energy mechanisms. Anglen et al.¹⁹ found that majority (31 out of 48) of patients had sustained high-energy injuries. Twenty-four out of fourty-eight patients had motor vehicular accidents. This difference in mechanism of injuries cannot be explained by difference in age of the patients in these studies.

3.3. Fracture characteristics

Table 3 shows distribution of fractures in different studies according to Letournel and Judet classification system. Most common fracture pattern was both column fracture, seen in 601/3157 patients (19.03%), followed by anterior column and posterior hemitransverse fracture in 544/3157 (17.23%), anterior column fractures in 541/3157 (17.13%) and posterior wall fractures in 425/3157 patients (13.46%). It can be seen that fracture patterns differ in elderly patients as compared to their younger counterparts. Elderly patients had a higher incidence of involvement of anterior column and quadrilateral plate, and impaction of the articular surface.^2,20 In younger patients, posterior wall fracture was the most common fracture pattern seen, followed by transverse with posterior wall, and T-type fractures.^21,22 Ferguson et al.¹ found that anterior column fracture, and anterior column with posterior hemitransverse fracture were more common among the geriatric patients as compared to the younger patients, and transverse with posterior wall fracture pattern was more common among the younger patients.

Table 3.

Distribution of fractures in different studies according to Letournel and Judet classification system.

Studies	Total	Posterior wall	Posterior Column	Anterior wall	Anterior column	Transverse	T type	Both column	Posterior wall + posterior column	Anterior column + posterior hemitransverse	Transverse + posterior wall	Others
Anglen et al. 200319	48	11	1	0	5	13	3	5	6	0	4	0
Archdeacon et al 201324	39	0	0	0	7	0	1	9	0	22	0	0
Bible et al 201420	86	17	0	2	14	5	5	12	2	16	13	0
Boelch et al. 201640	23	3	0	0	3	0	0	15	1	1	0	0
Boraiah et al. 200925	19	12	1	0	0	1	0	1	1	1	2	0
Borg et al. 201942	13	1	1	0	0	1	0	5	0	5	0	0
Carroll et al. 201021	93	15	5	2	6	2	7	26		20	10	0
Chakravarty et al. 201414	17	0	0	0	1	3	1	4	0	5	3	0
Deren et al. 201717	99	13	2	6	34	11	0	16	5	5	7	0
Enoscon et al. 201458	15	0	0	0	10	1	0	0	0	4	0	0
Ernstberger et al. 202030	608	40	26	94	191	29	33	47	4	123	12	9
Ferguson et al. 20101	235	31	1	8	45	2	24	62	8	35	19	0
Firoozabadi et al. 201631	156	18	2	8	0	10	10	58	0	50	0	0
Gary et al. 20127	35	0	1	0	7	2	5	13	0	5	2	0
Giunta et al. 201818	27	9	2	0	0	4	2	5	0	5	0	0
Herath et al 20192	96	5	2	0	33	2	1	18	2	29	4	0
Herscovici et al. 201047	22	0	0	0	0	0	0	6	0	7	9	0
Jeffcoat et al. 201248	41	0	0	0	1	0	1	19	0	20	0	0
Kim et al. 201549	186	23	6	21	34	8	21	42	8	14	9	0
Laflamme 201538	9	0	0	0	4	0	1	2	0	2	0	0
Li et al. 201450	46	5	1	8	5	4	6	15	1	1	0	0
Lin et al. 201551	34	14	0	2	0	0	0	2	4	5	7	0
Lont et al. 201952	59	3	0	0	6	1	35	6	3	3	2	0
Malhotra et al. 201353	15	3	3	0	0	2	0	0	3	2	2	0
Manson et al. 201759	269	46	11	6	40	12	12	59	20	40	21	2
Mears et al. 20028	57	11	3	0	10	10	3	4	6	6	4	0
Miller et al. 201060	45	9	0	4	1	2	2	11	2	11	3	0
Mitchell et al. 201822	146	44	5	9	13	7	4	20	5	18	21	0
Moushine et al. 200411	18	0	0	0	0	2	9	0	2	1	4	0
Otoole et al. 201461	46	17	1	0	4	1	3	7	7	2	4	0
Papadakos et al. 201427	71	6	3	2	11	4	9	3		25	2	6
Resch et al. 201745	30	1	0	0	4	8	5	4	0	5	0	3
Rickman et al. 201454	24	1	0	0	4	8	0	6	3	2	0	0
Rodrı´guez et al. 201256	6	1	0	0	1	1	0	0	1	1	1	0
Ryan et al. 201734	27	1	0	1	5	4	3	4	0	9	0	0
Salama et al. 201626	18	7	0	0	0	3	2	1	1	1	1	2
Schnaser et al 201632	171	30	2	0	2	10	17	58	18	18	15	0
Tidermark et al. 2003	10	0	0	0	0	5	0	0	0	5	0	0
Walley et al. 201735	86	7	0	9	27	12	1	10	8	9	3	0
Weaver et al. 201844	59	16	0	0	8	3	3	8	0	11	10	0
Zha et al. 201713	53	5	1	4	5	3	9	18	1	–	7	0
TOTAL	3157	425 (13.46%)	80 (2.53%)	186 (5.89%)	541 (17.13%)	196 (6.20%)	238 (7.53%)	601 (19.03%)	122 (3.86%)	544 (17.23%)	201 (6.36%)	23 (0.72%)

Interesting, it was seen that some authors with high-energy mechanism of injury in their patients had higher numbers of posterior fracture patterns.^1,15,19,20 Conversely, several studies with low-energy mechanism of injury had predominantly anterior fracture patterns.^14,26 Several authors who performed primary THA for acetabulum fractures also had higher proportions of posterior injuries.^8,18,27, 28, 29

3.4. Treatment options

There is no consensus on ideal treatment for geriatric acetabulum fractures, and they have been treated using different methods in different studies. Treatment can also be individualized according to the fracture pattern and health and fitness of the patients. Of the 5160 patients for whom details of treatment were available, 2199 (42.62%) were given nonoperative treatment, 2285 (44.28%) were treated with ORIF of acetabulum fracture, 161 (3.12%) were treated with percutaneous fixation and 515 (9.98%) were treated with primary THA. Distribution of studies in terms of patients in each group is summarized in Table 4.

Table 4.

Details of treatment and mortality rates in the included studies.

Studies	Total	ORIF	Percutaneous Fixation	Conservative	CHP (Primary THA)	Fracture management during primary THA	1 year mortality	Perioperative/in-hospital mortality	Reoperations
Anglen et al. 200319	48	48	0	0	0	na	3 (6.25%)	0	8 underwent THA
Archdeacon et al. 201324	39	39	0	0	0	na	10 (25.64%)	0	5 underwent THA at mean 18 months
Baker et al. 201916	49	0	0	49	0	na	12 (24%)	na	na
Beaule et al. 200446	10	0	0	0	10	ORIF of the fracture prior to THA using single anterior approach. Porous coated acetabular component	na	na	na
Bible et al. 201420	86	55	0	31	0	na	6 (6.98%)	1 (1.16%)	na
Boelch et al. 201640	32	23	0	0	9	4 patients with both column fractures: ORIF and THA using posterior approach. 5 patients, 1 anterior column, two transverse & 2 anterior wall, THA without ORIF. Antiprotrusio cage used in all cases	1 (#.12%)	na	ORIF:2 underwent THA (at 7, 5.3 months), 1 girdlestone (3.7 months)
Boraiah et al. 200925	18	0	0	0	18	Posterior approach. Displaced column fractures were reduced and stabilized with plate and screws. Pressfit cup.	na	na	1
Borg et al. 201942	13	0	0	0	13	ORIF with posterior approach in 5 and anterior approach in 8 patients. THA with posterior approach in 11 and anterior approach in 2 patients. Titanium Burch–Schneider ring used.	2 (15.38%)	0	9/14 in ORIF group underwent THA. 1/14 in ORIF underwent girdlestone.
Boudissa et al. 201933	82	19	0	44	19	ORIF with posterior approach and THA. Press-fit cup or a Kerboull crossplate	18 (21.95%)	na	4 (21%) of ORIF group had secondary THA over 2 years
Carroll et al. 201021	93	84	0	0	9	na	na	na	26 underwent THA over 5 years
Chakravarty et al. 201414	19	0	0	0	19	percutaneously placed cannulated screws. Posterior approach for THA. Press fit cup.	5 (26.32%)	2 (1.05%)	na
Deren et al. 201717	99	17	0	76	6	na	19 (19.19%)	8 (8.08%)	6/23 with ORIF underwent THA
Enoscon et al. 201458	15	0	0	0	15	Anterolateral modified Hardinge in 13 & posterolateral Moore in 2. Burch–Schneider reinforcement ring	3 (20%)	0	0
Ernstberger et al. 202030	608	117	62	429	0	na	na	na	na
Ferguson et al. 20101	235	na	na	na	na	na	na	na	na
Firoozabadi et al. 201631	156	57	0	99	0	na	Surgery 7 (12%); Non-surgical 44 (44%)	na	na
Gary et al. 20117	80	0	80	0	0	na	10 (12.5%)	1 (1.25%)	19 underwent THA
Giunta et al. 201818	27	0	0	0	27	Kerboull cross-plate was used for ORIF.4 patients modified stoppa approach and THA through Kocher-langenbeck approach, two patients, for 2 patients the outer window of the ilioinguinal approach in conjunction with the Stoppa approach was used. Cemented dual mobility cup.	0	2 (7.41%)	na
Guerado et al. 201247	4	0	0	0	4	3.5 mm DCP used for fixation. Anterolateral approach. Cemented cup.	1 (25%)	na	na
Herath et al. 20192	1914	967	0	947	0	na	na	101 (5.28%)	na
Herscovici et al. 201047	22	0	0	0	22	ORIF through Kocher-Langenbeck (KL) or ilioinguinal approach. Ganz ring acetabular component	na	na	5 revision surgery
Jeffcoat et al. 201248	41	41	0	0	0	na	na	na	11 underwent THA
Kim et al. 201549	186	94	10	81	1	na	14 (7.53%)	na	na
Kim et al. 202015	2471	na		0	na	na	na	na	na
Laflamme et al. 201538	9	3	0	3	9	na	0	0	3 underwent THA
Li et al. 201450	52	52	0	0	0	na	na	na	5 underwent THA
Lin et al. 201551	33	0		0	33	ORIF through 4 ilioinguinal and 28 kocher-langenback’s approach. Uncemented reflection cup in all except 1 cemented restoration cup.	6 (18.18%)	na	2 underwent THA
Lont et al. 201952	59	25	0	0	34	Anterior intrapelvic approach±ilioinguinal approach for anterior and lateral parts of pelvis. Kocher-Langenback’s approach for posterior part and THA. GAP II reinforcement ring.	ORIF 2.25 (9%); THA 3.74 (11%)	na	9 underwent THA
Malhotra et al. 201353	15	0	0	0	15	ORIF and THA through posterior approach. Octopus ring.	0	0	na
Manson et al. 201759	269	142	0	107	20	na	na	na	na
Mears et al. 20028	57	0	0	0	57	Posterolateral approach used for 16 patients, anterolateral used for 38 patients and extended lateral approach used for 3. Harris-Galante-I or II acetabular cup.	na	na	3 underwent THA
Miller et al. 201060	45	45	0	0	0	na	na	na	13 underwent THA
Mitchell et al. 201822	146	88	na	58	na	na	24 (16.6%)	na	na
Moushine et al. 200411	18	0	0	0	18	Cerclage wire through Kocher-Langenback approach. Press-fit cup with a superior flange and an inferior hook.	1 (5.56)	0	0
Ortega-Briones et al. 201763	24	0	0	0	24	ORIF by Kocher langenback and modified stoppa approach.	na	na	1 underwent THA
Otoole et al. 201461	61	52	0	9	0	na	0	0	13 underwent THA
Papadakos et al. 201427	71	22	0	43	5	na	na	na	na
Resch et al. 201745	30	0	0	0	30	watson-jones approach. Roof-Reinforcement Plate 3.5 + cemented cup.	na	na	nil
Rickman et al. 201454	24	0	0	0	24	modified stoppa for anterior fixation and kocher-langenback for posterior fixation and THA. Trabecular metal revision modular shell	3 (12.5%)	1 (4.17%)	1 underwent THA
Rodrı´guez et al. 201256	6	0	0	0	6	trabecular metal revision cup through posterior approach	0	0	1 underwent THA
Ryan et al. 201734	27	0	0	27	0	na	24%	na	4 underwent THA
Salama et al. 201626	18	0	0	0	18	ORIF via Kocher langenback in all except 1 ilioinguinal approach. Cementless acetabular cup	na	na	na
Schnaser et al. 201632	171	90	0	80	1	na	17 (9.9%)	na	4 underwent THA
Tidermark et al. 200357	10	0	0	0	10	Burch-Schneider Antiprotrusion Cage via anterolateral in 6 and posterior in 4	na	na	na
Walley et al. 201735	86	37	0	49	0	na	17 (19.77%)	na	6 underwent THA
Weaver et al. 201844	70	33	0	0	37	ORIF by Kocher langenback or illioinguinal approach	14 (20%)	na	15 (10 in ORIF and 5 in THA)
Wollmerstädt et al. 202,064	176	82	10	67/161 patients (41.6%)	2/61	na	25%	10 (5.7%)	20 THA (5 in Non-operative and 15 in Operative group)
Zha et al. 201713	53	53	0	0	0	na	2 (3.23%)	NA	2 underwent THA
TOTAL	7876	2285	161	2199	515

Note: The total of column 3,4,5&6 is not expected to match total in column number 2.

3.5. Conservative treatment

Many elderly patients with acetabulum fractures may be high risk surgical candidates. Their functional demands may be less. They may have poor bone stock making fracture fixation difficult. Since they are low energy injuries, proportion of undisplaced fractures may be high.^1,30 Undisplaced fractures can be managed conservatively. Certain fracture types like associated both column fractures with secondary congruence may also be managed conservatively. This fracture pattern was relatively more common in elderly as compared to younger patients.^1,30, 31, 32

Anterior fracture patterns are more likely to be managed conservatively than posterior fracture patterns.^33,34 In the study by Baker et al.,¹⁶ 49 patients were managed conservatively, and anterior column posterior hemi-transverse was seen in 41 patients (84%) of them. Similarly, Walley et al.³⁵ found that nonoperative treatment was most often performed for anterior-column fracture patterns.

Patients receiving conservative treatment were, in general, older than those receiving operative treatment. Schnaser et al.³² found that average age of the patients treated operatively was 69 years, and the average age of the patient treated nonoperatively was 73 years. This difference was statistically significant (P < 0.05). Data from German pelvic registry has shown that elderly patients with acetabulum fractures were significantly more likely to get conservative treatment as compared to their younger counterparts.² They did not see a significant difference in distribution of fracture types in patients treated nonoperatively compared to those treated operatively. Ernstberger et al.³⁰ found that patients receiving nonoperative treatment were significantly older than patients treated with surgery.

Presence of significant co-morbidities may also be a deciding factor for nonoperative treatment. Wollmerstädt et al.³⁶ showed that patients who were treated nonoperatively were more likely to have an ASA score >2 (56/67, 83.6%) when compared to patients who were treated operatively (75/109, 68.8%, p = 0.03). Giunta et al.¹⁸ compared 27 patients undergoing primary THA with 21 patients undergoing nonoperative treatment. Patients undergoing nonoperative treatment had significantly higher ASA grades, poorer functional outcomes and higher mortality rates.

It is not clear whether outcomes differ between operative and nonoperative treatments. Similar outcomes have been reported by several authors for operative as well nonoperative treatment of these fractures, particularly in anterior fracture patterns. Herath et al.² did not find any significant difference in quality of life score (EQ-5D) between the operatively and nonoperatively treated subgroups. Majority of them (41%) had poor reduction according to Matta et al.³⁷ at one year. Yet, no correlation was seen between quality of reduction and mobility status. Ryan et al.³⁴ reported good functional outcomes of displaced acetabular fractures treated nonoperatively, that should have been managed by operative treatment according to the fracture displacement. However, their population presents some selection bias because none of the patients had posteriorly unstable fracture pattern. Baker et al.¹⁶ studied 49 patients over 65 years of age who are considered unfit for surgery. They are treated conservatively. Only 25% of them were minimally displaced. No correlation was seen between fracture reduction and mobility status at follow-up. Walley et al.³⁵ compared clinical results between operative and nonoperative treatment of acetabular fractures in the elderly. They did not find a statistically significant difference in mortality at one year. In a study by Ernstberger et al.,³⁰ there was no effect of the type of treatment (nonoperative vs. operative) on mortality at one and two years (P = 0.65 and P = 0.10).

On the other hand, poorer functional outcomes of nonoperative treatment have also been reported. Boudissa et al.³³ studied displaced acetabular fractures in elderly. They were preferentially managed operatively. Indication of nonoperative treatment in displaced acetabular fractures was presence of co-morbidities. Functional outcomes were inferior in patients managed nonoperatively as compared to those who underwent surgery.

Elderly patients have limited functional demands. Operative treatment may not always achieve good articular congruency in these patients. This may partly explain these findings. Thus in elderly, low demand patient who are not surgical candidates due to associated co-morbidities, conservative treatment appears to be appropriate.

3.6. Open reduction and internal fixation (ORIF)

ORIF is the preferred treatment for displaced acetabulum fractures in younger patients. In elderly patients there are concerns regarding higher risk of perioperative complications and poorer bone quality. Walley et al.³⁵ found higher morbidity after these fractures treated operatively compared to conservative treatment. But operative treatment was not associated with increased mortality. Patients treated operatively had a significantly higher numbers of treatment failures and conversions to THA within one year.

Giannoudis et al.³⁹ in a metaanalysis on age unspecified acetabulum fractures found rates of osteoarthritis after fracture fixation to be 26.6%, and severe osteoarthritis was seen in 19.1% patients. A high failure rate of ORIF in elderly has been seen in literature, ranging from 0 to 33%.^{7,17,19,23,25,38} Not all fractures are at a high risk of failure of fixation and patterns associated with poorer outcomes should be identified before surgery is planned.

Schnaser et al.³² found a higher conversion to THA in patients managed with ORIF as compared to patients managed conservatively (3.8% vs 15%). Herath et al.² had a 24.7% conversion to THA secondary to osteoarthritis after fracture fixation. Rates of secondary THA in patients managed conservatively was 15.8%. The difference between the two groups was not statistically significant. About three-fourth of secondary hip replacements were necessary within the first 12 months after the fracture.²

Zha et al.¹³ divided 62 patients above 60 years undergoing surgery for acetabulum fractures into two groups: young–old group (60–74 years) and old–old group (75–90 years). Accurate reduction was attempted for the former and a limited operative procedure was used for the latter. Old-old group had higher pain scores and lower ambulation score but overall modified Postel Merle D’Aubigne Score was similar between two groups. Whereas average clinical scores were significantly higher in the young–old patients with anatomical reduction, no statistical difference in clinical scores was seen between the different grades of fracture reduction in old–old group. They concluded that the reduction quality in the old–old patients may be not affecting the clinical function.

Operated patients and nonoperatively treated patients are often not comparable in terms of fracture pattern, and it is difficult to generalize conclusions from these studies. Displaced fractures should be operated as far as possible and aim should be to achieve early mobilization. It should be understood that ORIF may not provide advantage of earlier mobilization as compared to conservative treatment.

3.7. Percutaneous fixation

This is an infrequently performed procedure for geriatric acetabulum fractures. Since displacement is expected to be less, these patients may be more amenable to percutaneous reduction and fixation. Gary et al.⁷ reported a series of 79 patients, 60 years of age or older, who underwent percutaneous reduction and fixation of the acetabulum. Twenty patients (25%) ultimately required THA with a mean time to THA of 1.41 years. Mouhsine et al.,⁴¹ studied 21 patients with a mean age of 81 years, with undisplaced or minimally displaced acetabular fractures, who were treated with percutaneous fixation. At a mean follow-up of 3.5 years, 17 of 18 patients had excellent or good clinical outcomes. No evidence of fracture displacement was seen. Ernstberger et al.³⁰ compared outcomes of ORIF, percutaneous fixation and nonoperative treatment. Percutaneous fixation was associated with lesser blood loss, shorter operative time, shorter duration of hospital stay and lesser nonsurgical complications compared to ORIF. No difference in quality of reduction was seen between percutaneous and open treatment.

Percutaneous fixation may have a role in treatment of acetabulum fractures in elderly. Careful patient selection is required for this procedure. Percutaneous fixation can also be used to stabilize the columns before a primary THA.¹⁴

3.8. Primary THA

Combined hip procedure refers to one-stage management of these injuries. 29 studies had used primary THA as a method of treatment in some or all of their patients. A total of 515 patients underwent primary THA in all the studies. Management of fractures during surgery in these patients is summarized in Table 4.

Primary THA in conjunction with ORIF would require posterior approach alone for fractures amenable to fixation using posterior approach.²⁵ Alternatively, two separate approaches may be required if anterior fracture line is reduced and fixed from anterior approach and THA is performed using a posterior approach. Percutaneous reduction of the anterior column can also be combined with THA.¹⁴ Extension of ilio-femoral approach can also be utilized for THA, if both the fracture fixation and THA are done anteriorly.⁴⁶ Fracture reduction can be easily performed as complete visualization of the acetabulum can be obtained after removal of femoral head. Once the femoral head is removed, deforming forces of the leg are also absent. Columns and walls can be secured with screws and plates at the time of fracture fixation.^27,40 Restoration of acetabular bone stock is a very important component of THA in acetabulum fractures. In cases of comminuted posterior wall fractures, where fractured bone cannot be reconstructed, the posterior wall can be reconstructed with a femoral head autograft.²⁷ The femoral head can also be used as a morselised bone graft to fill the defect due to protrusio deformity created as a result of displaced anterior column or quadrilateral plate fractures.^27,40,42

Failure of ORIF needing THA is well known in acetabulum fractures. Similar to the concept followed in proximal femoral fractures, treatment should be a safe and definitive surgical procedure, with low risk of re-surgery. Another important aim of treatment of elderly patients with acetabulum fractures is early mobilization. Identifying patients who might need an arthroplasty after ORIF is important. Patients with these poor prognostic factors for outcomes after fracture fixation should be offered primary arthroplasty. There is some evidence in literature that THA after a failed ORIF has poorer outcomes as compared to a primary THA.^4,5 In study by Weber et al.,⁴³ 27 out of 66 patients needed revision of the acetabular component at a mean follow up of 9.6 years.

Borg et al.⁴² compared similar acetabular fractures treated with either ORIF or with a combined hip procedure. In the ORIF group, nine of 14 patients required revision surgery to THA within two years of the index procedure. All these patients were elderly patients, presenting with comminuted acetabular fractures with severe acetabular impaction, with or without concomitant femoral head injury. Weaver et al.⁴⁴ found a higher rate of reoperation (10/33, 30%) with ORIF compared with those treated with THA (5/37, 14%); however, this was not statistically significant (P = 0.12).

Boelch et al.⁴⁰ found revision rates to be lower in THA. Failure rates of ORIF needing THA was 45%. Two out of nine patients in the THA group needed revision surgery.

Boudissa et al.³³ used primary THA for posterior fractures of acetabulum and ORIF for anterior fractures of acetabulum. This rationale cannot be applied as universal. Articular depression and femoral head injury are frequently seen in anterior fracture patterns. Gull sign is often seen in anterior fracture patterns and may benefit from primary THA. Mears⁹ recently reported 8-year outcomes of 57 patients treated with ORIF of the acetabulum vs primary THA. He concluded that the primary THA to be a promising option in selected cases of acetabular fractures.

Due consideration should be given to fracture characteristics, and weather it can be reduced anatomically. If fracture cannot be reduced well, primary THA may be a promising option in elderly patients with acetabulum fractures. Mobilization after primary THA for acetabulum fractures appears to be earlier as compared to ORIF. Boraiah et al.²⁷ used toe touch weight bearing for 8 weeks. Mears et al.⁸ and Borg et al.⁴² mobilized patients on crutches, full weight-bearing. Resch et al.⁴⁵ used mobilization protocol of full weight bearing if possible within the first 10 days. This appears to be an important benefit of primary THA.

Further studies with sufficient sample size are required to compare complication rates between ORIF and primary THA.

3.9. Indications for primary THA

Primary THA is an acceptable method for treatment of some of these injuries. Important challenge is to identify patients who will be better served by THA rather than internal fixation. Several radiological findings have been associated with poorer clinical outcomes. Zha et al.¹³ found that Gull sign, femoral head injury (wear, abrasions, or defects in the articular cartilage) and posterior wall comminution (more than three separate fragments of the posterior wall) were associated with poor quality of reduction and poorer outcomes. Posterior wall comminution was also associated with early loss of reduction and failure of fixation. They did not observe poorer clinical outcomes with marginal impaction of posterior wall and quadrilateral plate fractures. In patients with Gull sign, functional outcomes correlated with quality of reduction.

Anglen et al.¹⁹ found that superomedial dome impaction or Gull sign was a major predictor of failure of internal fixation. Anatomic reduction was closely related to good and excellent radiographic results. In series of Beaule et al.,⁴⁶ all patients who underwent THA had significant articular depression (>50% of the roof) of the acetabulum or an associated displaced femoral neck fracture. According to Boelch et al.,⁴⁰ the fracture pattern was the main criterion for selecting primary THA as the treatment of choice. Indications included comminuted fractures and associated femoral head fractures. Full-thickness cartilage loss on femur, impaction of the femoral head or impaction of the acetabulum involving >40% of the weight bearing area were indications of arthroplasty for Mears and Velyvis⁸. Boraiah et al.²⁷ had femoral head impaction in 10 patients out of total 15 patients with posterior wall fractures treated with THA.

A displaced quadrilateral plate fracture does not appear to be associated with poor outcomes. Archdeacon et al.²⁶ studied 26 patients (follow-up more than 12 months) with protrusio fractures of acetabulum. They could achieve good clinical and radiological outcomes in these patients. Reduction quality did not correlate well with the outcomes. About 50% of the patients had less than an anatomic reduction, but only 19% needed THA.

Posterior wall comminution and extension of fracture to weight bearing dome was associated with poorer outcomes in posterior wall fractures.^6,7

Carroll et al.²³ identified poor prognostic factors predicting need for THA after ORIF. 26 out of 93 patients in their series needed a THA following ORIF. They found that marginal impaction, fracture comminution, poor fracture reduction, associated femoral neck fracture, fractures irreducible through single non-extensile approach and protracted duration of surgery expected, significant femoral head impaction, severe osteoporosis and preexisting arthrosis were associated with poorer outcomes with ORIF and need for THA.

3.10. Rehabilitation

Early mobilization is one of the primary goals in managing lower limb injuries in elderly. It aims to avoid complications of prolonged recumbency. Traction is poorly tolerated by elderly patients. Elderly comply poorly with partial weight-bearing protocols and weight-bearing may be guided by limits of pain.

Return to baseline ambulation was overall poor after acetabulum fractures.^35,36 Seventy-six percent of operated patients did not return to their baseline ambulation by their latest follow-up in the study by Walley et al.³⁵ Wollmerstädt et al.³⁶ found that after this injury only 65% of the patients were able to eventually return to their homes. On the other hand, Anglen et al.¹⁹ found that 85% of the patients returned to their pre-operative status of community ambulation after the surgery. 52% of these patients used some assistive device for ambulation. 77% patients were completely pain-free. THA was associated with earlier mobilization as compared to ORIF.⁴⁰

3.11. Mortality

Elderly patients with trauma have high in-hospital mortality rates.⁴⁹ In-hospital mortality rates in elderly patients with acetabulum fracture has been seen to vary from 0 to 8%. One-year mortality ranged from 0 to 26%. Higher mortality rates of up to 30% at one-year have also been seen.²⁵ Mortality was significantly associated with age of the patients.^2,19 Herath et al.² found that the mortality in patients younger than 60 was 2.8% compared to 5.3% in those more than 60 years of age. Anglen et al.¹⁹ had average age of the patients who died after surgery as 77 years (range 61–87), which was significantly higher than rest of the patients. Chakravarty et al.¹⁴ found that patients with high-energy injuries had lower age (65 years) as compared to low-energy injuries (80 years). Mortality in the former was significantly lower as compared to the latter, which can be explained by difference in mean age.

Associated injuries may have a significant contribution to mortality in these patients. Bible et al.²⁰ divided patients into two groups: those with isolated acetabular injuries and those with acetabular injuries with other associated injuries. One-year Mortality rates were significantly different between the two groups (8.1% vs 23.3% at one year). The greatest difference in mortality rates between the two groups was seen in the first 30 days.

Schnaser et al.³² had 6.6% one-year mortality rate for patients who underwent surgery. Mortality for nonoperatively treated patients was 13.8%. Herath et al.² also found a higher mortality in nonoperatively treated patients. This may be because the age of patients in nonoperatively treated group was higher, and there might have been a selection bias against surgery for patients with co-morbidities. Mortality in elderly patients treated operatively was found to be lower than those treated nonoperatively by several authors.^20,32 Walley et al. ³⁵ did not find a statistically significant difference in mortality rates between the operative and nonoperative treatment of these injuries.

Sarcopenia, or age-related decrease in muscle mass, has also been associated with increased mortality. Mitchell et al.²⁴ found mortality rates of 32.4% in patients with sarcopenia, compared with 11.0% in patients without sarcopenia. Deren et al.¹⁷ studied 192 patients above 60 years of age with acetabulum fractures. 42.4% of them had sarcopenia. Sarcopenia was significantly associated (p = 0.0419) with increased one-year mortality (28.6%) compared with the absence of sarcopenia (12.3%).

4. Discussion

Acetabulum fractures in elderly is a growing problem. Herath et al.² studied 15-year data from the German Pelvic Registry and found that geriatric patients accounted for 50.5% of all acetabulum fractures. Fergusson et al.¹ studied a database of 1309 acetabulum fractures enrolled over 27 years (1980–2007) and found the number of elderly patients with acetabulum fractures to constantly increase. It is difficult to describe who should be categorized as geriatric patients. Included studies were heterogenous in age cut-offs for geriatric acetabulum fractures (Table 1). Cut off of 55 years was selected as several studies had defined elderly patients as those above 55 years of age.

Low injury mechanisms are predominant in geriatric fractures. Difference in fracture pattern in elderly patients can be explained by the difference in mechanism of injury in those patients. Though there is no study correlating mechanism of injury with pattern of fracture, it can be postulated that fall in these patients produces an impact on the greater trochanter which fractures anterior column and pushes it superiorly, and displaces the quadrilateral plate medially. There may be articular depression at the edge of the fracture and impression fracture on the head of femur.³

Hip fractures in elderly are common and have received a major attention. Similarly, acetabulum fractures in elderly have a major impact on mobility and longevity of elderly patients and need more attention. Although, mortality rates of acetabular fractures treated operatively was lower than mortality of hip fractures treated operatively in elderly,²⁰ they cannot be considered to be an injury to neglect. Unsatisfactory results have been reported in about one-third of these patients treated nonoperatively.^14,15 Often age, comorbidities and presence of osteoporosis dissuade the surgeon from operative treatment, even in those fractures which would otherwise have been treated operatively in younger patients.³

Reported outcomes of ORIF for acetabular fractures are variable. Outcomes of ORIF in acetabulum fractures in elderly was considered to be unsatisfactory by several authors.^10,11 But many studies have shown satisfactory outcomes of ORIF depending on the quality of reduction achieved.¹² Irrespective of the age, rate of conversion to secondary THA after ORIF was 19.8%.¹³ This was comparable to the rates found in different studies in this review. Fragility fractures of hip and spine have been found to be associated with increased one-year mortality. Acetabulum fractures in elderly also had a high one-year mortality.

Huge differences have been seen in proportions of geriatric fracture treated operatively in different institutions. Thus, the choice of treatment appears to be based on choice of the surgeon and prevailing practices rather than on evidence. Studies included in this review are retrospective and heterogenous in respect to patient population, fracture characteristics, treatment given and outcomes measured. The comparison groups, if present are not uniform in respect to patients and fracture types. Selection of treatment chosen has been arbitrary and non-uniform. For example, patients with less displaced fractures, severe co-morbidities and extreme age were often chosen for nonoperative treatment in most series. Clear criteria for division of patients in different treatment arms was mentioned in only a few studies. Studies did not mention clinical outcomes according to fracture types. Primary THA was often carried out for fractures not considered to be amenable to fixation by the operating team. Most studies had small sample size and were at best retrospective case series. Hence data from these studies cannot be combined for further analysis. But a useful glimpse of current knowledge on geriatric acetabulum fracture is presented to highlight lacunae in current knowledge.

Literature on treatment of geriatric acetabulum fractures is not enough to draw any definite conclusions. There is limited evidence from current literature that surgery could be considered a safe option for treatment of displaced acetabulum fractures in elderly. Primary THA can provide early mobility and reduce chances of resurgery in fracture patterns where restoration of joint surface may not be possible.

Declaration of competing interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.