Surgical outcomes of subtypes of periprosthetic tibia fractures after total knee arthroplasty

Jiayong Liu; Josh Vander Maten; Julia Beyer; Logan J. Roebke; Muhammad Z. Moral; Nabil A. Ebraheim

doi:10.1016/j.jor.2021.12.003

View/Download PDF

Buy Reprints

PDF

Translate this page into:

29 (); 11-14

doi:

10.1016/j.jor.2021.12.003

Surgical outcomes of subtypes of periprosthetic tibia fractures after total knee arthroplasty

Jiayong Liu^a,, Josh Vander Maten^b, Julia Beyer^b, Logan J. Roebke^b, Muhammad Z. Moral^a, Nabil A. Ebraheim^a

a University of Toledo Medical Center, Department of Orthopedic Surgery, Toledo, OH, 43614, United States

b University of Toledo, College of Medicine and Life Sciences, Toledo, OH, 43614, United States

∗Corresponding author: Jiayong Liu. jiayong.liu@utoledo.edu

Received: 2021-12-4, Accepted: 2021-12-24, Published: 2021-12-28

Professor P K Surendran Memorial Education Foundation

Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Surgical outcomes of subtypes of periprosthetic tibia fractures after total knee arthroplasty were evaluated by using the Felix et al. classification system. Type 3 fractures were the most common classification of periprosthetic tibial fractures. Type 2 fractures had the highest rates of revision and nonunion. Type 3 fractures exhibited longer healing times than types 2 and 4. Far type 3 fractures showed the longest healing time of all fracture types but had very minimal complications. Type 4 fracture managed by K-wire/cerclage wire may require hardware removal or debridement but exhibited the shortest healing time compared to types 2 and 3.

Keywords

Periprosthetic tibia fracture

Total knee arthroplasty

Tibia

TKA

Felix et al. classification system

Show Related Articles from PubMed

1 Introduction

Total knee arthroplasty (TKA) procedures performed are on the rise. As the frequency of the procedure increases, there is a greater chance of a surgeon having to deal with associated complications.^1–10 One complex complication that can arise is a periprosthetic tibial fracture (PTF). This is a rare complication with an occurrence rate of less than 1%.²^,³^,^8–11

Current literature has few examples of studies involving total knee arthroplasty with periprosthetic tibial fracture. One study, by Felix et al.,¹² established a classification for such cases. The classification system categorizes fractures into both a type (1, 2, 3,4) and a subclass (A, B, C). Type 1 fractures partially cross the tibial head. Type 2 fractures extend fully across the tibial head. Type 3 fractures cross the tibia distal to the stem component of the prosthesis, and type 4 fractures involve the tibial tubercle. Subclassifications include A indicating a stable prosthesis, B indicating an unstable prosthesis, and C indicating the fracture occurred intraoperatively.¹² This classification was utilized to categorize fractures in this study [Fig. 1].

Illustration of the Felix et al. classification system of periprosthetic tibia fractures. Anatomic location (1-Plateau; 2-Adjacent to stem; 3-Distal to stem; 4-Tibial tubercle) + Subtype (A-Well-fixed prosthesis; B-Loose prosthesis; C-Intraoperative) = Fracture Type.

Although a classification system for TKAs with PTF has been established, surgical outcomes of subtypes of periprosthetic tibia fractures after total knee arthroplasty have not been thoroughly evaluated. This study at a level 1 trauma center will analyze surgical outcomes of subtypes of periprosthetic tibia fractures after total knee arthroplasty.

2 Methods

This retrospective study analyzed all tibia procedures from 2007 to 2020 at a level 1 trauma center. Institutional review board approval was obtained for the retrospective review of patient medical records and radiographs. Patient demographics of age, sex, comorbidities, mechanism of injury, and fracture side were reviewed.

Only patients with periprosthetic tibial fractures following a total knee arthroplasty were included in the study. Once potential patients were identified, a more in-depth chart review was used to analyze demographics, standard of care, and complications.

Based on radiograph files, patients were then categorized based on fracture type and subclass using the classification system established by Felix et al.¹² In addition to the classification system, a special denotation was made if the fracture occurred extremely distal to the prosthesis tip near the ankle region. A comprehensive analysis of the patient characteristics and standard of care was then compiled. Chart review data was then analyzed using the Excel Data Analysis function. Outcomes were measured based on healing time, revision rate, and complication rate in each subtype group.

3 Results

A total of 24 patients were included in this study. 12 females and 12 males were included in the study with an average age of 71.79 ± 10.70 (53-92) years. Fracture classifications were as follows: Ten patients had 3A fractures, seven 3A(far), three 2A, two 2B, one 4A, and one 2C fracture. Twenty-three fractures occurred postoperatively, and 1 occurred intraoperatively.

In two patients, bone grafts were used as determined necessary by the attending surgeon. The patient cohort had a high prevalence of comorbidities with many having predisposing conditions that could significantly impact the bone quality. Fifteen of the 24 patients had a comorbidity that predisposed them to poorer bone quality; Nine patients had osteoarthritis, three had osteoporosis, and two had rheumatoid arthritis [Table 1]. Other conditions are listed in Table 1.

Table 1 Patient demographics and comorbidities.

Case	Age	Gender	BMI	Classification	Bone Graft (Y/N)	Significant Diseases
1	56	F	29	2B	N	HTN, Osteochondroma (shoulder)
2	85	F	21.7	3A (far)	N	Osteoarthritis
3	76	M	36	2C	N	Osteoarthritis
4	81	F	35.4	2A	N	Osteoporosis
5	66	M	31.4	2B	N	Non-ambulatory for 3+ years
6	85	F	24.3	3A	N	N/A
7	87	F	22.3	2A	Y	Osteoporosis, Osteoarthritis
8	65	M	37.7	3A	N	N/A
9	54	M	31.9	3A (far)	N	N/A
10	71	M	32.1	3A	N	Osteomyelitis
11	53	F	55.8	4A	N	N/A
12	64	M	28.7	3A (far)	N	N/A
13	64	M	26.6	2A	N	N/A
14	86	F	43.3	3A (far)	N	Osteoarthritis
15	68	F	41.3	3A (far)	N	Rheumatoid Arthritis
16	66	M	35.2	3A	Y	Rheumatoid Arthritis
17	82	M	40.3	3A	N	Osteoarthritis
18	71	M	32.7	3A	N	Osteoarthritis
19	69	F	34.4	3A (far)	N	N/A
20	66	M	39.6	3A	N	Osteoarthritis
21	71	M	N/A	3A	N	N/A
22	76	F	32.4	3A	N	Osteoarthritis
23	69	F	42.8	3A	N	Osteoarthritis, Osteoporosis
24	92	F	31.8	3A (far)	N	Cancer

BMI is a confounding variable that could potentially predispose TKA patients to PTF. In combination with the high number of comorbid bone conditions in this cohort, many of these patients were predisposed to the fracture they eventually developed. With the exclusion of one patient whose BMI could not be identified, the average BMI for all patients was 34.20 ± 7.68 (21.7–55.8) indicating obesity or a grade II BMI (30–39.9 kg/m). Six patients had a grade III, or severely/morbidly obese, BMI (≥40 kg/m). Twelve individual patients had grade II while the remaining patients included three overweight, or grade I, patients, and only three with an optimal weight.

The most common mechanism of injury was a fall, occurring in 18 of the patients. Other mechanisms of injury included two crush injuries, two hit by car, one intraoperative, and one power wheelchair accident. Conservative treatment was attempted in four patients. Of note, eventual surgical management was necessary in all of the conservatively treated patients. The lone malunion case was managed initially by attempted conservative treatment.

External fixation and locking plate was by far the most commonly used surgical technique [Fig. 2]. A variety of other surgical approaches were also used based on surgeon preference and individual patient needs [Table 2]. With the exclusion of the patients lost to follow up, death, or intraoperative cases, only two fractures resulted in nonunion. Of the 16 patients that achieved union, the average healing time was 110.63 ± 57.35 (48-282) days. The average time from arthroplasty to PTF was 262.58 ± 272.86 (1-842) days and the average time from injury to surgery was 16.13 ± 47.30 (0-234) days.

A-D: 64 y/o M, had a car crash injury, Type 2 & 3 fractures, treated with external fixator, A- AP view, B-Lateral view. The fractures were healed post-OP 4 months, the external fixator was removed, C-AP view, D-lateral view. E-F: 85 y/o F, had a fall on the ground, Type 2 A fracture, E-Pre-OP AP view, F-Pre-OP lateral view, G-AP view post-OP 4 months, It showed the fracture healed, H- Lateral view post-OP 4 months, It showed the fracture healed.

Table 2 Patient treatment and outcome data.

Case	Union time (days)	Conservative treatment	Type of Treatment	Complications	Time from Arthroplasty to Fracture (days)	Time from injury to surgery (days)
1	Nonunion	N	External fixator-- > plate-- > external fixator	Infection, revised to locking plate and then TKA removal	1	234
2	68	Y	External fixator	Malunion	142.86	16
3	Intraoperative	Y	Antibiotic Spacer-- > long stem	Revised with cannulated screw	85.29	42
4	74	Y	Non-operative splinting	Revised with locking plate	263.43	14
5	65	N	External fixator	none	63.14	0
6	77	N	Locking plate and screw	none	270	1
7	61	N	Locking plate and screw	2 bone infusions after nonunion	53.29	1
8	114	N	Locking plate and screw	none	5.57	3
9	137	N	External fixator	none	177.14	9
10	138	N	External fixator	Refracture case; Pin site infection w/removal	529.14	12
11	73	N	K-wire cerclage wire	Stiffness w/debridement surgery; Symptomatic hardware removal surgery	33	17
12	92	N	External fixator	none	131.71	1
13	147	N	External fixator	Revision surgery of the wound - Skin graft surgery	51.86	2
14	282	N	External fixator	none	768.86	7
15	10	N	Locking plate and screw	none	-	1
16	48	N	Locking plate and screw	Compartment syndrome + Fasciotomy	586.14	1
17	162	N	External fixator + External locking plate	Compartment syndrome + Fasciotomy	-	1
18	73	Y	External locking plate	none	153.14	2
19	99	N	External fixator	none	162.71	1
20	128	N	External locking plate	Compartment syndrome on presentation; Skin graft surgery	842	5
21	Died	N	Locking plate and screw	none	-	4
22	Died	N	External locking plate	Revised with external fixator	-	10
23	Died	N	Locking plate and screw	none	668.71	2
24	Died	N	External fixator	none	-	1

When examined based on fracture classification, the 3 patients with 2A fractures showed an average union time of 94 days. One patient had nonunion and received 2 bone infusions. One patient required wound revision and skin graft. There were 2 patients with 2B fractures, with an average union time of 65 days. Revision surgery was required in one patient. There was one case of nonunion and one of infection. There was one patient with 2C fracture. This patient initially received an antibiotic spacer then long stem implant, which was later revised to cannulated screw. There were 10 patients with 3A fractures with an average union time of 106 days. One revision surgery was required and complications included one refracture, one pin site infection, and three cases of compartment syndrome. There were 7 patients with 3A (far) fractures with average healing time of 114 days. The only complication recorded in this group was malunion in a patient initially managed nonoperatively before requiring external fixation. The single patient with a 4A fracture exhibited a union time of 73 days. This patient did not require revision, but experienced stiffness and required debridement and removal of symptomatic hardware.

4 Discussion

Periprosthetic tibia fractures following total knee arthroplasty remain a complex and uncommon complication. The patient cohort included in this study consisted of a population with a high prevalence of comorbidities and a well above average BMI increasing the difficulty of management. Management of these fractures is highly variable, but union can be achieved through surgical means. Surgical management was the preferred method of treatment at this level 1 trauma center. With an average union of time of 110.63 ± 57.35 (48-282) days, surgical management provides satisfactory results for PTF. Of note, conservative treatment was attempted in four cases and included the lone malunion. A common complication associated with PTF in this case series was compartment syndrome which presented either before or after the surgery.

Due to the rapidly increasing number of TKAs performed annually, the incidence of PTF is expected to increase as well (2-10). Pannu et al. report that there are fewer than 200 cases of PTF in the current literature in 2021(2). Because of the rarity of these fractures, there is no consensus on optimal management, though the expected increase in incidence makes this a pressing issue. Many papers in the current literature focus on outcomes of PTF, some of which report based on Felix et al. classification,²^,⁷^,⁹^,¹⁰^,¹²^,¹³ but time to union is rarely reported. Sim et al. report union time in their study, but only for type 2A fractures.³ We attempt to address this gap in the literature by reporting union time for each included classification.

While type 1 fractures and/or fractures of the tibial plateau have been reported as the most common fracture type,⁶^,⁸^,⁹^,¹²^,¹⁴ we found type 3 fractures to be most common in our study. The far type 3 fractures exhibited the longest healing time, which may be due to disrupted blood supply to the distal portion, as blood flow runs proximal to distal. However, these fractures showed the least number of complications, consistent with Schreiner et al. who reported the highest 5 year survival rate of implants in type 3 fractures.⁹ In our study, type 2 fractures had the highest rates of revision and nonunion, which is consistent with reports in the current literature.⁴^,⁷^,¹³^,¹⁴

All our patients initially managed conservatively eventually required surgical intervention. Some advocate for nonoperative treatment of fractures that are stable and nondisplaced with a well-fixed implant.⁴^,⁶^,¹⁰ However, surgical management is much more frequently employed⁷^,¹⁰^,¹³^,¹⁴ and allows earlier weight-bearing while avoiding possible complications of conservative management such as deep vein thrombosis and pressure sores.⁴ Limitations of this study include its small sample size and retrospective design. More data is needed to develop standardized treatment recommendations for PTF and further prospective randomized studies will aid in reaching a consensus on management of these rare fractures.

5 Conclusions

In conclusion, type 3 fractures are the most common classification of periprosthetic tibial fractures. Locking plate and external fixation are the most commonly used treatment methods. Type 2 fractures had the highest rates of revision and nonunion. Type 3 fractures exhibited longer healing times than types 2 and 4. They were associated with the highest rates of compartment syndrome. Far type 3 fractures showed the longest healing time of all fracture types, but had very minimal complications. Type 4 fracture managed by K-wire/cerclage wire may require hardware removal or debridement but exhibited the shortest healing time compared to types 2 and 3.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References

Chimutengwende-Gordon M., Khan W., Johnstone D., . Recent advances and developments in knee surgery: principles of periprosthetic knee fracture management. Open Orthop J. 2012;6:301-304.
[Google Scholar]
Pannu T.S., Villa J.M., Cohen E.M., Hayda R.A., Higuera C.A., Deren M.E., . Periprosthetic tibial fractures after total knee arthroplasty: early and long-term clinical outcomes. J Arthroplasty. 2021 Apr;36(4):1429-1436.
[Google Scholar]
Sim J.A., Shon O.J., Joo Y.B., Sohn H.S., Byun S.E., Kim J.W., . Clinical outcomes of osteosynthesis of well-fixed periprosthetic proximal tibial fractures (Felix type 2A) after total knee arthroplasty. Injury. 2021 Oct;52(10):3099-3103.
[Google Scholar]
Mohler S.A., Stambough J.B., Mears S.C., Barnes C.L., Stronach B.M., . A review of periprosthetic tibial fractures: diagnosis and treatment. Orthop Clin N Am. 2021 Oct;52(4):357-368.
[Google Scholar]
Damsgaard C.W., Gad B.V., Bono O.J., et al . Intraoperative proximal tibia periprosthetic fractures in primary total knee arthroplasty. J Knee Surg. 2021 Oct;34(12):1269-1274.
[Google Scholar]
Born C.T., Gil J.A., Johnson J.P., . Periprosthetic tibial fractures. J Am Acad Orthop Surg. 2018 Apr 15;26(8):e167-e172.
[Google Scholar]
Morwood M.P., Gebhart S.S., Zamith N., Mir H.R., . Outcomes of fixation for periprosthetic tibia fractures around and below total knee arthroplasty. Injury. 2019 Apr;50(4):978-982.
[Google Scholar]
Ebraheim N.A., Ray J.R., Wandtke M.E., Buchanan G.S., Sanford C.G., Liu J., . Systematic review of periprosthetic tibia fracture after total knee arthroplasties. World J Orthoped. 2015 Sep 18;6(8):649-654.
[Google Scholar]
Schreiner A.J., Schmidutz F., Ateschrang A., et al . Periprosthetic tibial fractures in total knee arthroplasty - an outcome analysis of a challenging and underreported surgical issue. BMC Muscoskel Disord. 2018 Sep 11;19(1):323.
[Google Scholar]
Platzer P., Schuster R., Aldrian S., et al . Management and outcome of periprosthetic fractures after total knee arthroplasty. J Trauma. 2010 Jun;68(6):1464-1470.
[Google Scholar]
Agarwal S., Sharma R.K., Jain J.K., . Periprosthetic fractures after total knee arthroplasty. J Orthop Surg. 2014;22(1):24-29.
[Google Scholar]
Felix N.A., Stuart M.J., Hanssen A.D., . Periprosthetic fractures of the tibia associated with total knee arthroplasty. Clin Orthop Relat Res (345):113-124.
[Google Scholar]
Bauer C., Zaharia B., Galliot F., et al . Management and results in periprosthetic tibial fracture after total knee arthroplasty: two-center 15-case retrospective series at 2 years' follow-up. Orthop Traumatol Surg Res. 2020 May;106(3):449-458.
[Google Scholar]
Cacciola G., Mancino F., De Meo F., Bruschetta A., De Martino I., Cavaliere P., . Current reconstruction options in periprosthetic fractures around the knee. Geriatr Orthop Surg Rehabil. 2021 Aug 27;12
[Google Scholar]