Use of INFIX for managing unstable anterior pelvic ring injuries: A systematic review

Abstract

Introduction

Subcutaneous screw rod system which is popularly known as Pelvic internal fixator (INFIX) has emerged as an alternative to external fixators in management of unstable pelvic injuries. INFIX has shown various advantages over external fixation such as reduced infection rate and patient morbidity. However, it has its own set of complications such as lateral femoral cutaneous nerve injury, heterotopic ossification, femoral nerve palsy etc. We intended to conduct a systematic review of the current literature to assess outcomes and complications with INFIX technique of fixation.

Methods

A comprehensive search of literature was performed based on Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines and online database of EMBASE, PubMed, Medline and Scopus was searched for all studies in English language till March 2020. Included studies were reviewed for demographic data, fracture type/classification, radiological outcome and functional outcomes. The inclusion criteria were: 1. Studies in English language 2. Clinical studies reporting use of INFIX technique in pelvis fracture where clinical and radiological outcomes were reported.

Results

Twenty-two studies fulfilling inclusion and exclusion criteria were included in this systematic review with total of 619 patients. Radiographic parameters and outcome measures were infrequently reported. Fixation with INFIX in these fractures leads to 87% excellent to good radiological results and 84% excellent to good functional results. Complications include lateral femoral cutaneous nerve irritation (25.3%), heterotopic ossification (24.7%), infection (3%), and femoral nerve palsy (1.6%); which is likely related to placing the bar and screws too deep.

Conclusion

This analysis supports the use of INFIX in management of unstable pelvis fractures where anterior fixation is required.

1. Introduction

Pelvic injuries make up between 2 and 8% of all fractures of the human body but they are associated with significant mortality and morbidity. Stable fractures can be treated non-operatively; however unstable fractures require surgical intervention.^{1, 2, 3, 4, 5}

Unstable fracture patterns requiring surgical intervention may need posterior, anterior or combined fixation methods depending on fracture pattern.^5,6 Anterior pelvic ring fixation can be achieved with open reduction and internal fixation (ORIF) or external fixation.^5,6 ORIF is associated with prolonged operative time and morbidity, while external fixators are associated with high complication rates.^{5, 6, 7, 8} Subcutaneous screw rod system which is popularly known as Pelvic internal fixator (INFIX) has emerged as an alternative to external fixator.⁹ INFIX has shown various advantages over external fixation such as reduced infection rate and patient morbidity. However, it has its own set of complications such as lateral femoral cutaneous nerve injury, heterotopic ossification, femoral nerve palsy etc.^5,9 A systematic review on INFIX was published in 2018, however multiple new studies have been published since then and hence the need for an updated review. Proper analysis of this modality of treatment is not available leading to uncertainties in the choice of treatment for anterior fixation. We intended to conduct a systematic review of the current literature to assess outcomes and complications with INFIX technique of fixation.

2. Methods

A comprehensive search of literature was performed based on Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines as on March 2020. The online database of EMBASE, PubMed, Medline and Scopus were searched. Our aim was to retrieve all clinical studies of English language including prospective and retrospective studies reporting on the results of INFIX technique in pelvic fractures. The keywords used were “INFIX” “pelvic fracture” “external fixator” and “anterior subcutaneous internal pelvic fixation”. Furthermore, manual search of reference studies of all included studies was also searched for any additional relevant studies. The review was submitted for PROSPERO registration (ID 190382). Quality of studies was assessed by GRADE.¹⁰

2.1. Study selection and data extraction

All possible relevant studies were retrieved from the online database. The titles were screened for relevance to INFIX technique in pelvic fractures. Subsequently the articles were selected and their abstracts were retrieved. Full texts of selected articles were extracted. The inclusion criteria were: 1. Studies in English language 2. Clinical studies reporting use of INFIX technique in pelvis fracture where Clinical and radiological outcomes were reported. Case reports, case series with less than 4 cases, letter to editor, biomechanical studies, review articles and cadaveric studies were excluded. The search was performed by two authors (CK and SM) independently; who then reviewed references of the qualifying papers and selected appropriate studies on the basis of inclusion and exclusion criteria. The two investigators then independently reviewed the following data: demographic data, fracture type/classification, radiological outcome and functional outcomes. Third author (KK) was consulted in case of any disagreement between the two authors. (Fig. 1)

Fig. 1.

Study selection and Data extraction as per PRISMA protocol.

Characteristics of the studies^{11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32} included are as per Table 1.

Table 1.

Characteristics and demographic profile of included studies.

Study	Study design	Country	Sample size	Mean age (years)	Follow-Up (months)	Quality
Yasin(2020) 11	Prosspective	Egypt	16	29.5	–	Moderate
Hua X(2019) 12	Retrospective	Germany	23	37.6	6	Low
Li(2019) 13	Retrospective	China	28	42	20.1	High
Steer(2019) 14	Retrospective & Prospective	Australia	24	38.5	Not reported	Moderate
Yin(2019) 15	Retrospective	China	35	41.7	27	High
Vaidya et al., 2012 16	Retrospective	USA	91	39.2	15	Moderate
Gardner et al., 2012 17	Retrospective	USA	24	39	12	Low
Merriman et al., 2012 18	Retrospective	USA	13		7.25	Low
Muller et al., 2013 19	Retrospective	Germany	36	45.2	54	Low
Scheyerer et 2014 20	Retrospective	Switzerland	4	65	Not reported	Low
Hesse et al., 2014 21	Retrospective		8		6	Low
Hoskins et al., 2016 22	Prospective	Australia	21	39	11.4	Low
Zhang et al., 2016 23	Retrospective	China	28		8.7	Low
Gang et al., 2016 24	Retrospective	China	12		<6	Low
Zhaohui et al., 2016 25	Retrospective	China	12		7.5	Low
Wu et al., 2018 26	Retrospective	China	23		15	Moderate
Fang et al., 2017 27	Retrospective	China, Germany and switzerland	29	64.2	7.2	High
Bi et al., 2017 28	Retrospective	China	21	37.8	16.6	High
Vaidya et al., 2017 29	Retrospective	USA	83	41.67	33.29	High
wang et al., 2017 30	Retrospective	China	26	39.3	6	High
Shetty et al., 2017 31	Prospective	India	15	37.7	34.9	High
Dahill et al., 2017 32	Prospective	UK	47	41	12	Moderate

2.2. Quality assessment

The quality of all studies were assessed independently by two authors (C.K and S.M) using Grading of Recommendations Assessment, Development and Evaluation (GRADE) recommendations.¹⁰ Any discrepancy between the two reviewers was resolved by third reviewer (KK).

3. Results

Twenty-two studies fulfilling inclusion and exclusion criteria were included in this systematic review. (11–32) There were a total of 619 patients. The mean age was 42.2 years (range 29.5–64.2). In 531/619 patients 317 were male and 214 were females.

14 studies used OTA classification, while 9 studies used Young and Burgess Classification. Six studies mentioned both classifications, 3 studies used Tile classification while 2 studies didn’t elaborate on the classification used. In patients who were classified using OTA classification, 208 were OTA61C and 239 were OTA 61B. The mean follow-up was 17.29months (range 4.2–54 months) (see Table 2).

Table 2.

Injury type, functional outcome, radiological outcome & complications in included studies.

Studies	Injury types									Functional outcome (Majeed)				Radiological Outcome				Complications
	OTA 61C	OTA 61B	YB APC	YB LC	YB VS	YB CM	Tile A	Tile B	Tile C	Excellent	Good	Fair	Poor	Excellent	Good	Fair	Poor	Heterotopic ossification	LFCN	Infection
Yasin et al., 202011								10	6	15	1	0		13	2	1	0	2	3	0
Hua X et al., 2019 12								6	22					15	9	4		8	6	0
Li et al., 201913								13	10	13	6	4		14	6	2	1	0	2	0
Steer et al., 201914			2	10	7	5				IOWA (5)	(7)	(2)	(1)					5	11	2
Yin et al., 201915	10	25												7	19	6	3	12	10	1
Vaidya et al.,201216	41	50	30		20	7												32	27	3
Gardner et al., 201217	24																	6	2	1
Merriman et al., 201218																				1
Muller et al., 201319	36									GERMAN (9)	(11)	(8)	(3)					9	6	2
Scheyerer et al.,201420	2	2	1	2		1												1	0	0
Hesse et al., 2014 21	6	2	1	3	2	1												NR	NR
Hoskins et al., 201622			3	18														9	12	3
Zhang et al.,201623	6	22	3	19	6														12	0
Gang et al., 2016 24	3	9	4	5	3													0	0	0
Zhaohui et al., 2016 25	8	4																0	4	0
Wu et al., 201826	5	18								14	7	2		12	8	3		8	3	0
Fang et al., 2017 27	15	14																	14	1
Bi et al., 2017 28		21								10	9	2							3	0
Vaidya et al., 201729	39	44	20	39	18	6												56	7	3
Wang et al., 2017 30		26	1							21		5						0	2	0
Shetty et al., 201731	13	2								8	7			12	3			0	1	1
Dahill et al., 2017 32																		0	26	1

Surgical technique: Surgical steps and fixation method used varied slightly among studies but basic steps remained same as described by Vaidya et al.¹⁶ with few modifications in some studies. Most surgeons used mini-open or percutaneous methods to put screws under strict fluoroscopic guidance. The implants used were pedicle screws and rods of various spine pedicle screw fixation systems. Some studies used polyaxial while other preferred monoaxial screws. First supraacetabular screws were put on each side. Then the rod is contoured and put subcutaneously at the level of the bikini line. After reducing fracture by different indirect mechanisms and checking reduction under image guide, this rod is fixed with screws. The length and thickness of screws varied from study to study and patient to patient, usually bulkier patients requiring longer screws. Screws used were 5.5–9 mm; most common is 6.5–7 mm in diameter and 55–120 mm in length depending on the size of the patients and bone quality.^11,12 Rods used were 5–6.5 mm in diameter. Few studies used modified INFIX technique with three pedicle screws, where a small third screw was used in pubic tubercle.^26,28,30 Posterior fixation was usually attempted first wherever required. Patient’s ambulation was allowed from as early as tolerated in some studies to 4–6 weeks in others and it was also dependent on injury. Elective implant removal was advised and done in-between 12 weeks to 9 months which required a second surgery in an operation theater. All fractures healed except 2 Nonunion reported in 2 studies; one each by Muller et al.¹⁹ and Dahill et al.³²

Radiological outcomes: Radiological outcomes were reported by 8 studies only (mostly by Matta outcome scores). Out of 164 Patients reported, 85(51.8%) had excellent, 58(35.4%) had good,13(7.9%) had fair and 8(4.9%) had poor outcomes. This indicates that fixation with INFIX in these fractures leads to 87% excellent to good radiological results. Functional outcome: nine studies reported functional outcome scores with different methods which included 7 as per Majeed score, 1 by German outcome and 1 by IOWA pelvic score. Out of 251 Patients reported, 120(47.8%) had excellent,91(36.3%) had good,39(15.5%) had fair and 1(0.4%) had poor outcome.

Lateral femoral cutaneous nerve (LFCN)Injury: The lateral femoral cutaneous nerve (LFCN) injury or irritation was reported to be most common complication. In this analysis LCFN injury was present in 151 of 598 reported patients (25.3%) and it was persistent in 36 patients (6%).It presented as paresthesia or numb patch on anterolateral thigh. Seventeen studies found that the LFCN injury was temporary and in most cases, it improved with removal of the implant.^{16,17,22,23,25,26,29,30} Four studies found it to be a persistent issue.^14,19,27,32 Of these four, Steer et al.¹⁴ reported LFCN injury in 11/24 patients, two bilateral. 7/48 (14.6%) of these patients resolved after removal of implant but in six patients (25%), symptoms were persisting at their most recent follow-up, after 6 months of implant removal. Study by Muller et al.¹⁹ found that 6/31 patients (19.4%) reported persistent symptoms (unilateral in four and bilateral in two cases) at the time of follow-up. These symptoms were persistent but not significantly physically disturbing and manifested mainly as hypoesthesia, not as allodynia, hyperpathia, or causalgia. Fang et al.²⁷ reported 14/29 (48.3%) injuries to the LFCN, with six presenting as minor paresthesia and eight (27.5%) as painful neuropathic symptoms which were persistent. Dahil et al.³² reported 26 patients (55%) had injury of the LFCN, nine of which were bilateral. Ten of these patients recovered fully but 16 (34%) were having persistent symptoms.

Femoral nerve palsy: Five studies reported femoral nerve palsy in 10 patients.^{21,25, 26, 27, 28} The study by Hesse et al.²¹ is mainly about femoral nerve palsy after the use of INFIX. They reported 6 patients who had 8 femoral nerve palsies. In one patient it was recognized early, and in 5 patients, it was recognized late. All patients were treated with INFIX removal. Only 2 of 8 femoral nerves recovered fully and rest were not recovered fully by the last follow-up. The authors believed that it was due to deep interconnecting rod which had led to compression of the femoral nerve. The other 4 studies reported 1 case each. The treatment remained removal^25,27,28 or readjustment²⁶ of INFIX. Three cases recovered fully, and 1 partially. Although it is a rare complication but it is a devastating one, so the surgical steps for INFIX application should be followed meticulously; and placing the screws and bar too deep should be avoided.

Heterotopic ossification (HO) was another very common reported complication. It was reported in 148 of 598(24.7%) cases but it was almost always asymptomatic except in 2 cases in the study of Vaidya et al.²⁹ One patient who had severe HO declined removal and second patient had severe HO unrelated to the INFIX. In this study, older age and higher injury severity score were found to be associated with high chances of developing HO while length of time the INFIX was in place, Glasgow Coma Scale, sex, race, and Majeed functional score showed no significant relationship with the development of HO.²⁹ HO was usually asymptomatic and required no treatment.

Infection was reported in 19 of 619(3.1%) cases.^{16,18,19,22,27,29,31,32} Initially it was treated with antibiotics and local measures such as irrigation but it required implant removal in 11 cases along with culture specific antibiotics.^14,16,19,29 Infections resolved in all cases, with no reports of developing osteomyelitis.^{6,18,19,22,27,29,31,32}

Implant failure: Implant failure was described wherever implant was removed before the proper fracture healing. Early reoperations were required in 35/619 cases. Early reoperation was required due to loss of fixation in 10 patients(failure of fixation was at the screw rode connection) and due to infection in 9. Rest of the early reoperation, 16/619 were likely caused by the screws and bar placed too deep which caused compression of the underlying structures, resulting in discomfort(5patients), femoral nerve palsy(10 patients), and LFCN irritation (1 patient). It is important to place the screws and rod above the fascia so as to prevent compression on the underlying structures.^{14,16,19,25,27, 28, 29}

4. Discussion

This analysis supports the use of INFIX in management of unstable pelvis fractures where anterior fixation is required. Although a relatively new technique, after initial learning curve, this technique is reproducible among various level of Orthopedic surgeon. However, one does need an experienced radiology technician to show appropriate intraoperative images, which may help surgeon to decrease the operative time.¹⁴

INFIX technique has various advantages. One of the advantages is that it can be used in conjunction with laparotomy in unstable pelvis fracture and in associated with acetabular fractures.³³ This technique is more appropriate in obese patient, urethral injury and minimally displaced fractures.³⁴

Studies comparing INFIX with symphyseal plating have demonstrated a better anatomic reduction with symphyseal plating (p.019)¹⁵; this may be because INFIX performs closed reduction of both upper and lower public rami fractures and with plating it is openly reduced. However, symphyseal plating leads to more blood loss (up to 4.5 times) as compared to INFIX. This is because of the fact that plating requires exposing the fracture site and this whole process of peeling of soft tissue and dissection around the area which is rich in blood vessel leads to more intraoperative bleeding.¹⁵ When compared with External fixator INFIX has fewer complications specifically of infection and implant loosening.²⁸ These good clinical results are corroborated with similar results in biomechanical studies.^35,36 INFIX can be used as definitive fixation in anterior pelvic injury. Apart from the high complication rate, external fixator limits patient’s daily activities such as sitting, lying down in lateral position, prone position, rolling over and sexual intercourse.^6,7

In our analysis, almost all fractures united except two.^19,32 Nonunion is a rare event in pelvic fractures as pelvis has a good vascular supply which helps in good healing if fractures are reduced adequately. So INFIX provide good fixation in combination with posterior fixation or alone.

Radiological and functional outcomes (with known specific parameters) were reported by very few studies. Objective Radiological parameters were mainly assessed by the Matta technique, and that too was used only postoperatively. Fixation with INFIX in these fractures leads to 87% excellent to good radiological results. The functional outcome was reported with use of Majeed, IOWA & German outcome scores. Functional outcomes showed that 84.1% patients in this analysis had excellent to good outcome. This correlates well with other studies evaluating INFIX.^{12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35} Although proper evaluation requires detailed assessment of individual score.

This new technique Of INFIX has its own set of unique complications. These are heterotopic ossification, LFCN & Femoral Nerve palsy. In our meta-analysis, LFCN palsy is the most common complication, reported in 151/598 patients (25.3%). It has been reported in most of the studies. It usually improves with time, once the implant is removed. It persisted in total 36/598(6%) patients reported by four studies.^14,19,27,32 Patients with persistent symptoms ranged from 19.4% to 34% by different studies which can be worrisome. Although no clear solution has been found how to decrease chances of LFCN injury but this technique should be followed with meticulous dissection to decrease it. Vaidya et al.¹⁶ recommended blunt dissection between Sartorius and Tensor fascia lata and gentle handling of LFCN in order to prevent its palsy. In spite of the best efforts by surgeons and following all precautions, some of the patients still end up having LFCN palsy. LFCN is approximately 2.2 cm away from the pedicle screws of INFIX, which explains the high incidence of LFCN injury with this technique.^36,37 The larger pedicle screw diameter has also been found/postulated to be a factor responsible for traction LFCN palsy.³⁶ This factor may be mitigated to some extent with use of smaller diameter (6.5 mm) screws. However, use of smaller diameter screws comes at the price of comprising stability of fixation.¹⁵ Future biomechanical and cadaveric studies need to be conducted to find the optimal diameter of pedicle screws.

Femoral nerve palsy is a rare complication but it is a devastating one.^{21,25, 26, 27, 28} It can be minimized with use of proper technique for INFIX application; and placing screws and bar too deep should be avoided. Osterhoff et al. showed that optimal rod to bone distance of 20 mm, followed by 30 mm to avoid complications caused due to compression.³⁷ So, It is best to keep the rod to bone distance of 20–25 mm, since it has the lowest rate of early removal and other complication due to compression of femoral neurovascular bundle.^35,37

Heterotopic ossification (HO) is another common complication of INFIX. Majority of these ossifications are asymptomatic without any effect on hip function. Older age and higher injury severity score were found to be associated with high chances of developing HO while the length of time the INFIX was in place, Glasgow Coma Scale, sex, race and Majeed functional score showed no significant relationship with the development of HO.²⁹ HO was usually asymptomatic and required no treatment. It can be prevented by thorough lavage of surgical site.¹⁶ Prophylactic treatment with radiation therapy or non-steroid anti-inflammatory drugs (NSAIDs) may also help.

INFIX is not immune to infection and infection was reported in 3.1% cases which are comparable to other surgical procedures. Early diagnosis and treatment with antibiotics usually lead to infection control. The reported infections are very less than infections reported with use of external fixators.^7,8 There were 10 reports of implant failure which required re-exploration. Proper technique and good quality implant are required to decrease this complication.

The ASIF/INFIX is an innovative surgical method for anterior pelvic ring fractures. Most of the studies reported that patients tolerate it well with good functional results and there are few complications if proper precautions are not taken while applying the screws and rod. But this procedure generally requires surgical implant removal as a second surgery; which can be done safely as a day care procedure.

This study has number of limitations. First, majority of the study included in the analysis were retrospective case series with no comparative group or randomized controlled trials with other fixation techniques was available. Second, there is variability in age group and also follow ups. Third, only a few studies had reported radiological and functional results in normally used objective scoring methods and rest simply observed subjective methods for assessment which limits our conclusion.

5. Conclusion

In conclusion, this systematic review confirms that INFIX is a reliable technique for anterior pelvic injury with excellent to good functional outcomes and limited complications. Further high-quality comparative studies with long term follow up are needed to strengthen the evidence of this novel technique.

Authors’ contributions

VT, SM &KK conceptualized the review.CK and SM reviewed the literature. CK, SM & KK wrote the manuscript where VT gave the critical inputs. All authors read the manuscript and agreed on the content.

Source of funding

None.

Declaration of competing interest

The authors declare that they have no competing interests.