Ilizarov External Frame Technique for Pirogoff Amputations With Ankle Disarticulation and Tibiocalcaneal Fusion

Methods

Twenty-four patients were retrospectively identified via a review of the medical records at the authors’ institution dating back to January 2004. The demographic data included the age and sex of the patient, cause of amputation, medical comorbidities (eg, diabetes, arteriosclerosis, and polyneuropathy), active inflammation, number of prior operations, and nicotine use. The circulation of the hindfoot and heel pad was preoperatively evaluated and documented using Doppler sonography of the tibialis posterior artery and the ankle-brachial index. The threshold for an adequate arterial brachial index was considered to be 0.5, a value that has been proposed by Pinzur et al ¹⁷ for Syme amputations. In cases in which the tibialis posterior artery was not palpable or located by Doppler, contrast arteriography was required to show good collateral circulation. A neurological consultation was requested for all of the patients to evaluate the neurological status, especially of the heel pad. The contraindications for the procedure were calcaneus or heel pad destruction, osteomyelitis of the calcaneus, an insensate heel pad, or inadequate blood supply at the ankle level without sufficient collateral circulation to the heel pad. The complications were recorded and were classified as minor (treated nonoperatively) or major (treated operatively), according to the method used by Katsenis et al ⁹ in a study of Ilizarov arthrodesis.

A total of 24 patients who received a Pirogoff amputation with Ilizarov external frame use between January 2004 and August 2011 were identified. The mean age of the 19 male and 5 female patients was 57.4 years (range, 29-76 years). The most common indication was gangrene with uncontrollable infection, chronic ulceration, and necrosis limited to the forefoot and midfoot due to diabetes mellitus with a concomitant vascular disorder (Figure 1; 16 patients). Four patients suffered from chronic osteomyelitis of the midfoot and/or the talus (Figure 2A and B). Four patients had sustained a crush injury with irreversible soft tissue damage and bone loss distal to the Chopart joint (Figure 3). An average of 4 operative procedures (range, 0-8) had been performed in other hospitals prior to the Pirogoff amputation. The procedures were minor toe amputations, debridement of infected gangrenous tissue, and Lisfranc/Chopart amputations in 4 patients.

OPEN IN VIEWER

Figure 1.

Magnetic resonance image (A) and radiographs (B, C) (patient 21 with a Charcot arthropathy) show an ulceration of the plantar skin, osteolysis of the fifth metatarsal head, and a luxation at the Chopart joint line. Clinical photograph 3.5 months postoperatively at the day of frame removal (D). Anteroposterior radiograph (E) and clinical photograph (F) 18 months after frame removal.

OPEN IN VIEWER

Figure 2.

Clinical photographs and intraoperative radiographs of patient 24 at the day of admission (A) with a chronic osteomyelitis of the midfoot and a soft tissue defect following abscess incision and intraoperatively showing the surgical incision (B). After amputation of the forefoot at the Chopart joint line, parallel K-wires in the distal tibia mark the resection level for the saw cut (C, D). The rotated calcaneus is fixed temporarily to the distal tibia with two 2-mm K-wires (E). The postoperative photograph shows the applied Ilizarov frame (F).

OPEN IN VIEWER

Figure 3.

Clinical photographs of patient 16 with a degloving injury of the foot but an intact heel pad.

Operative Technique

A fish-mouth-shaped surgical incision running to the dorsal aspect of the foot was started approximately 1 cm distal to the talonavicular joint and angled posteriorly to both the medial and lateral malleoli (Figure 2B). The incision was then continued vertically or obliquely around the plantar aspect of the foot to the level of the calcaneocuboid joint. On the dorsal aspect of the foot, the extensor tendons were severed, the dorsalis pedis artery was ligated, and the superficial and deep peroneal nerves were severed. As the plantar incision was deepened, the plantar fascia, ligaments, and flexor tendons were cut. The tibialis posterior artery and the tibial and sural nerves were severed and ligated, respectively. After adequate dissection of the skin flaps from the bone, the forefoot was disarticulated at the Chopart joint; then the talus and the forefoot were disarticulated and removed (Figure 2C). The distal tibia and fibula were cut perpendicular to the mechanical axis of the tibia with an oscillating saw. The calcaneus was then cut 2 to 3 cm proximal to the calcaneocuboid joint approximately 90 degrees to the plantar surface. The calcaneus was then rotated and connected to the distal tibia. This position was temporarily fixed with two 2-mm Kirschner wires that were distally inserted through the calcaneus and into the distal tibia.

After skin closure, the Ilizarov frame was fixed to the leg. The Ilizarov external fixators consisted of 3 rings: 2 tibial rings proximal to the arthrodesis and 1 calcaneal ring (Figures 1D and 2F). The 2 proximal rings were mounted with 2-mm wires and/or 6-mm half-pins to the tibia. Three wires with stoppers were inserted into the calcaneus. All of the wires were tensioned with the Ilizarov tensioning device to 110 kg.

During the first postoperative days, a custom-made weight-bearing platform was attached beneath the distal ring (Figure 4). This arrangement allowed immediate mobilization and pain-limited weight-bearing, even in cases with prolonged or complicated wound healing. After inpatient treatment for a mean of 24 days (range, 8-46 days), the patients were regularly reviewed every 2 weeks. At these visits, gradual compression between the tibial and calcaneal ring was applied until radiographs showed consolidation of the arthrodesis (Figure 5).

OPEN IN VIEWER

Figure 4.

Clinical photograph (patient 24) 3 weeks postoperatively with the applied weight-bearing platform (A) and anteroposterior (B) and lateral (C) plain radiographs.

OPEN IN VIEWER

Figure 5.

Anteroposterior (A) and lateral radiographs (B) 6 months after frame removal, showing complete consolidation of the tibiocalcaneal arthrodesis and a soft tissue thickness of the heel pad of 2 cm (patient 24).

Results

An initially well-healed Pirogoff stump was successfully achieved in 21 of the 24 patients (87.5%) (Table 1, Figures 6 and 7). The mean external fixation time (EFT) was 18.1 weeks (range, 12.7-26.6 weeks). The mean EFT was longer in the patients with peripheral polyneuropathy (PNP) than in the patients without PNP: 18.4 weeks (range, 12.9-26.8 weeks) versus 17.3 weeks (range, 12.7-20.1 weeks). After frame removal, clamshell prostheses were supplied by the hospital’s orthopedic workshop (Figure 6).

OPEN IN VIEWER

Table 1.

Clinical Details With Recorded Risk Factors, External Fixation Time, and Complications of the 24 Patients ^a

Patient No.	Age, y	Sex	Side	Indication for Amputation/Causative Disease	Ankle-Brachial Index <0.5	PNP	Nicotine Abuse	Active Inflammation	Prior Operations	EFT, d	Minor Complications	Major Complications	Late Complications/Died During Follow-Up
1 a	63	M	R	COM		X	X	X	7	139	PI
2	56	M	R	DM		X	X	X	8	127	PI, SSI	Wire breakage	TTA, died
3 a	33	M	R	Trauma					5	89		Wire breakage
4 a	55	M	L	DM	X	X	X	X	1	99	PI	SSI
5 a	29	M	R	COM				X	1	110
6	63	F	R	COM	X			X	6	122	PI	SSI	Died
7 a	49	M	L	Trauma			X		3	131	PI
8	61	M	L	DM	X	X	X					SSI-TTA	Died
9 a	68	M	R	DM		X			7	122	PI
10	67	F	L	DM		X	X	X	3	154	SSI		Died
11	68	F	L	DM	X	X		X	4	159			TTA; died
12 a	45	F	L	DM		X		X		90	PI
13	57	M	R	DM		X			7	107	PI	Wire breakage
14 a	73	M	R	DM		X		X	3	186	PI	Wire breakage
15	60	F	R	DM		X			3	108
16 a	48	M	L	Trauma					1	133	PI
17	74	M	R	DM	X	X	X	X	3			SSI-TFA	Died
18 a	46	M	L	DM		X			6	122		Wound edge necrosis
19 a	70	M	R	DM		X			4	172
20	60	M	R	DM		X	X	X	3	168	PI		Died
21 a	71	M	R	DM		X	X		5	104	PI
22 a	32	M	L	Trauma		X	X		0	98
23 a	76	M	L	DM	X	X	X		4	106		Heel pad ulceration, TTA
24 a	53	M	R	COM			X	X	4	141		Wire breakage

Abbreviations: COM, chronic osteomyelitis; DM, diabetes mellitus; EFT, external fixation time; F, female; L, left; M, male; PI, pin track infection; PNP, polyneuropathy; R, right; SSI, surgical site infection; TTA, transtibial amputation; TFA, transfemoral amputation.

a

Patients included in the Taniguchi score.

OPEN IN VIEWER

Figure 6.

Radiograph (A) and clinical photographs (B, C) of patient 16 at 36 months after frame removal with the custom-made clamshell prosthesis (C).

OPEN IN VIEWER

Figure 7.

Clinical photographs of patient 24 at 6 months after frame removal.

Soft tissue and wound-healing complications occurred in 8 patients. In 2 cases (patients 8 and 17), a higher level reamputation was necessary within the early postoperative days (at days 8 and 14, respectively) due to progressive and ascending infection. Three patients (patients 4, 6, and 23) needed revision surgery due to deep surgical site infections around the tibiocalcaneal arthrodesis (TCA). In 2 of these patients (patients 4 and 6), definitive soft tissue healing was ultimately successful after frame removal. Patient 23 had a persistent heel pad ulceration after frame removal and bony consolidation of the TCA. This patient declined any revision surgery, such as a revision arthrodesis or Syme amputation, but received a transtibial amputation. In 2 cases (patients 2 and 10), removing the stitches led to secondary wound healing without the need for revision surgery. One patient (patient 18) developed persistent wound edge necrosis that required local revision at the time of frame removal. The wound healed within 4 weeks without subsequent complications.

The frame-associated minor complications that could be treated nonoperatively consisted of pin track infections, which were documented in 12 patients. The treatments included local skin incisions at the pin sites and oral antibiotics for 2 weeks. Six patients with recurring infections required continuous low-dose oral antibiotic treatment while the frame was in place. Frame-associated major complications (wire breakage) that required additional procedures were found in 5 patients.

Two patients with initially well-healed Pirogoff stumps needed a transtibial amputation due to newly developed deep gangrene of the heel pad at 26 months (patient 2) and 18 months (patient 11) after the Pirogoff amputation. Patient 2, who had latent polyneuropathy with almost normal plantar sensation at the time of the initial Pirogoff amputation, developed a completely insensate heel pad. Patient 11 returned to our clinic with a complete vascular occlusion on lower limb angiography.

The Taniguchi scores are provided in Table 2. The mean follow-up for the 15 included patients was 44.9 months (range, 12-86 months). Overall, 67% of these patients achieved good or excellent results. Fair (27%) and poor (6%) results were only observed in the diabetic patient group.

OPEN IN VIEWER

Table 2.

Follow-Up Evaluation Using the Taniguchi Score

		Score
Patient-No.	Follow-up, mo	Pain	Function	Radiograph	Leg Length	Indoor Activity	Total Score	Evaluation
1	86	50	14	10	0	3	77	Good
3	84	40	14	6	3	3	66	Good
4	79	40	8	6	3	3	60	Good
5	76	55	18	10	3	3	89	Excellent
7	66	50	18	10	3	3	84	Excellent
9	50	30	10	6	3	0	49	Fair
12	50	60	20	8	3	3	94	Excellent
14	38	30	8	6	3	3	50	Fair
16	36	60	20	10	3	5	98	Excellent
18	25	30	14	8	3	3	58	Fair
19	21	25	12	8	3	3	51	Fair
21	18	45	14	10	3	3	75	Good
22	18	50	16	10	5	3	84	Excellent
23	14	15	6	6	0	0	27	Poor
24	12	55	20	10	3	3	91	Excellent

Discussion

Since diabetic vascular disorders and infectious complications account for most lower extremity amputations, ²⁴ the initial soft tissue and bone quality is unfavorable for a Pirogoff amputation with a TCA in many cases and accounts for the high failure rates.^16,21 In similar clinical situations and indications, such as tibiotalar or tibiocalcaneal arthrodesis in Charcot feet and infected ankle joints, external fixation and the Ilizarov external frame have proven to be reliable for limb salvage.^{1,4,7-9,19,23} The overall success rate of 87.5% in the current study is comparable to the above-described studies. The advantages of the Ilizarov frame are based on the minimally invasive, thin, transosseous wires that provide high biomechanical stability while avoiding the problems of internal fixation. ⁴ The gradual compression at the arthrodesis site maintains additionally stability, even under weight-bearing. ⁷ Because only a few studies (most with small sample sizes)^{3,10,13,16,18,21,22} or case reports^2,12 have described the Pirogoff disarticulation and because all of these studies used different internal or external fixation techniques, including K-wires, cannulated screws, and mono- or bilateral fixators, comparing results is difficult. One important distinction from the Ilizarov technique is the need for a certain period of immobilization without weight-bearing. The internal procedures require cast-immobilization for several weeks with an additional risk of ulceration due to pressure sores caused by the cast.^12,22 To our knowledge, only one other study has examined use of the Ilizarov frame for Pirogoff amputations. Einsiedel et al ³ used the Ilizarov frame in 6 patients—4 with diabetes, 1 with vascular disease, and 1 with chronic polyarthritis—and achieved a success rate of 67%. Two patients needed reamputations due to secondary wound infections. ³ In a modification of the Einsiedel et al ³ study, we used a weight-bearing platform at the distal ring that allowed early postoperative mobilization without compromising the stump. Control of and care for the soft tissues are possible at all treatment stages. The frame removal can be performed on an outpatient basis. Weight-bearing pain caused by the internal fixation material, which may require hardware removal, ¹¹ does not occur with the Ilizarov frame.

Nonunion of TCA has been described as one of the main reasons for Pirogoff amputation failure and higher level reamputations.^3,10,13,16 With the Ilizarov frame, we achieved TCA union in all of the patients in the current study (except patients 8 and 17, who needed early higher level reamputations), and the treatment was successful even in cases of prolonged or failed wound healing. Soft tissue healing was found to be the limiting factor for a successful Pirogoff amputation, especially in patients with diabetes and vascular compromise. This observation is consistent with the literature on other ankle disarticulation techniques and other distal amputations, in which an increasing risk of wound breakdown has been observed in cases of concomitant vascular disorders or diabetes.^16,17,24 Taniguchi et al ²¹ described wound complications requiring a transtibial amputation in 4 of 7 Pirogoff amputations caused by diabetes or occlusive vascular disorders. The authors suggested that the Pirogoff procedure might not be suitable for patients with vascular disease. Other authors have suggested using an ankle-brachial index threshold of 0.5 to ensure adequate vascular inflow for Pirogoff ¹² and Syme ¹⁷ disarticulations. Inadequate blood supply was the main factor predicting wound failure and the need for surgical revision in our patients, which is consistent with the findings of Pinzur et al ¹⁷ for Syme disarticulations. If diabetic patients meet the criteria for adequate blood supply, however, the Pirogoff procedure can be a limb salvage procedure with a high success rate, comparable to that of the Syme procedure. ¹⁷ As in other studies, the best functional outcomes were observed in the patients who required amputations due to trauma.^2,21 These patients are normally younger and do not have concomitant diseases and mobility limitations before their injuries. The opportunity to walk indoors without prosthetic support was a major advantage that most patients mentioned at the follow-up, independent of the initial indication for amputation.

There are procedure-specific complications and disadvantages. The cumbersome Ilizarov ring fixators are less comfortable for patients than are internal fixation devices or monolateral or bilateral fixators. The long average external fixation time (18.1 weeks) was comparable to that reported in other studies using the Ilizarov external frame for arthrodeses of neuropathic and infected ankle joints.^3,4,7,23 The reason for the high wire breakage rate may be explained by the indirect loading of the wires caused by using a weight-bearing platform. The mechanical stress on the wires when using a weight-bearing platform is up to 400% greater than when using direct weight bearing with foot-to-ground contact. ⁶ Although the wire exchange can usually be performed on an outpatient basis, it is still a major complication that increases the external fixation times; without early treatment, it can result in secondary stump deformities and delayed or incomplete union due to the compression loss at the arthrodesis site. Pin track infections were frequently observed in the present study, which is consistent with the rate described in the literature.^4,8,19 Although the course of these infections is usually uncomplicated, they can result in major complications, including skin and bone infections. To our knowledge there are no studies comparing the functional results of Syme and Pirogoff disarticulations. Therefore, no conclusions can be made whether the potential advantages of a preserved heel pad and the minimal loss of leg length warrant the additional risk of external or internal fixation needed for the TCA. A prospective outcome study comparing the 2 ankle disarticulation procedures would be necessary to be able to answer these questions.

Conclusion

The Pirogoff ankle disarticulation was a limb salvage procedure with overall good functional results that allowed easy and effective prosthetic fitting. Better functional outcomes may be expected in patients with traumatic foot injuries or osteomyelitis of the midfoot than in patients with a medical history of diabetes. We believe that the Ilizarov external frame is a reasonable alternative fixation method as it enables early weight-bearing, a high rate of TCA union, and constant control of soft tissues, particularly when combined with a weight-bearing platform. However, the external-frame-specific disadvantages and restrictions in vascular compromised patients should be considered.