A n k l e Pain F o l l o w i n g Total K n e e A r t h r o p l a s t y A Case Report J a m e s K. W e a v e r , M D , a n d R o b e r t S. D e r k a s h , M D
Abstract: The authors present their experience with four cases of ankle pain occurring in the immediate postoperative period after total knee arthroplasty. In a 40-month period, the authors performed 87 total knee arthroplasties. In the first 28 cases, an extramedullary alignment rod was used and there were no instances of ankle pain (0%). In the next 30 cases, an intramedullary alignment rod was employed and there were three cases of moderate to severe ankle pain (10%). In the last 29 cases, the medullary canal was decompressed, irrigated, and aspirated prior to insertion of the alignment rod. In this group there was only one case of mild ankle discomfort (3.5 %). An extensive investigation was made into one of the cases in which the intramedullary alignment rod was used but the canal was not aspirated. Information from this case investigation, observation of the three subsequent cases, and the notation of a decrease in the incidence of ankle pain once aspiration was routinely used are reported. The unproven hypothesis that ankle pain may result from compression of the marrow contents into the bone interstices is presented for consideration. Three of the four cases of ankle pain reported here were spontaneously resolved within 9 months after surgery. Key words: ankle pain, intramedullary alignment, marrow disruption, intramedullary pressure, knee arthroplasty, complications, ankle pain, conservative care.
In this case we address the troublesome complication of ankle pain following total knee arthroplasty (TKA). We report our evaluation of our initial case and discuss three subsequent similar cases of ankle pain.
the right knee was performed, again using an uncemented Miller-Galante implant (Fig. 1). An intramedullary rod was used without decompression. The procedure was expeditiously accomplished without complication, and the patient was discharged 4 days after surgery on t o u c h - d o w n weight bearing using crutches. Two weeks after surgery she had her sutures removed. She progressed well with therapy (range of motion, 00-95 ° ) and had minimal knee pain, but soon began to complain of ankle and supramalleolar pain. There was no swelling or calf tenderness that would suggest thrombophlebitis. At the time of her next visit, 5 weeks after surgery, her knee had progressed well but her ankle pain had become more severe. It was intended that progressive weight bearing would begin at that time, but she was unable to bear weight because of her ankle pain. Routine radiographs of the ankle were normal (Fig. 2). At that time she was afebrile with no swelling, but the ankle and supramalleolar area remained tender.
Case Reports Index Case A 57-year-old w o m a n with seronegative rheumatoid arthritis had previous hip arthroplasty, forefoot reconstruction, contralateral left knee arthroplasty with an uncemented Miller-Galante (Zimmer, Warsaw, IN) prosthesis 7 months earlier; all procedures were without difficulty. In May 1989 arthroplasty of From the University of Colorado Health Sciences Center and Orthopaedic Associates of Aspen and Glenwood, Glenwood Springs, Colorado.
Reprint requests: James K. Weaver, MD, Orthopaedic Associates, 2001 Blake Avenue, Glenwood Springs, CO 81601.
447
448
The Journal of Arthroplasty Vol. 7 Supplement 1992
Fig. I. (A, B) Postoperative radiographs of the right knee of a 57-year-old woman with serologically negated rheumatoid arthritis (index case). A cementless first-generation Miller-Galante knee arthroplasty was performed without complication.
Ankle m o t i o n was moderately painful. There was no discoloration, hypesthesia; or temperature changes that w o u l d suggest a reflex sympathetic dystrophy. Because of concern that she m a y have developed a deep infection, blood tests were obtained. They were
normal except for a slightly elevated alkaline phosphatase. At 8 weeks after surgery, the ankle pain had not improved. The patient could not bear weight and was becoming increasingly concerned. A b o n e scan was
Fig. 2. (A) Anteroposterior and (B) lateral radiographs of the right ankle 6 weeks following TKA. At this time the patient was complaining of severe ankle pain and was unable to begin weight bearing. No abnormality is seen on the routine radiographs.
Ankle Pain After TKA
Fig. 3. Technetium bone scan of the right tibia and ankle 3 months following TKA on the ipsilateral right knee. Diffuse, intense uptake is seen throughout the entire distal portion of the tibia.
•
Weaver and Derkash
449
l
obtained to exclude the possibility of a stress fracture. The scan displayed a diffuse, intense, and increased uptake t h r o u g h o u t the distal half of the right tibia (Fig. 3). Since the bone scan did not show the focal appearance typical of a stress fracture, concern was raised about the possibility of deep sepsis or even neoplasm. At that point a magnetic resonance image was obtained (Fig. 4), which showed diffuse changes on the T-weighted image and a vaguely cylindrical outline similar to the configuration of the
intramedullary alignment rod. At 12 weeks after s u p gery, the patient continued to exhibit symptoms, was still unable to bear weight, and was very concerned about the possibility of infection or neoplasm. A Craig trocar biopsy was obtained from the medullary canal of the distal tibia, revealing n o r m a l b o n e with no inflammation or evidence of neoplasm. With this information, the patient was willing to accept our explanation that her ankle pain and related symptoms were due to the mechanical intrusion of the m a r r o w contents into the interstices of the endosteum, which was disturbing the endosteal nerve supply. Four m o n t h s after surgery she was able to begin increasing her weight bearing, and by 9 m o n t h s she was completely pain-free and had returned to her active lifestyle.
Subsequent Cases
Fig. 4. MRI T2-weighted anterioposterior projections of both tibia. Note abnormality of marrow contents on right (arrow). There is an indistinct rectangular outline similar to the cross-sectional shape of the intramedullary alignment rod.
Experience with the index case sensitized us to the possibility of ankle pain following TKA and stimulated our interest in seeking possible solutions to the problem. Three additional cases of ankle pain following TKA are identified (Table 1), t h o u g h n o n e was as severe as the index case. All four cases were analyzed with attention to the technique of implantation and were compared to the entire group. Case 2 concerned a 78-year-old w o m a n w h o h a d a knee revision for loose components. The intramedullary rod was used without decompression. Her chief postoperative complaint was ankle and shin pain. With the experience we gained f r o m the previous case, we were able to reassure her, and indeed the pain disappeared approximately 6 m o n t h s after surgery. Case 3 was a 45-year-old m a n w h o had a TKA for severe traumatic arthritis. The intramedullary rod was used without decompression. Distal tibial and
450
The Journal of Arthroplasty Vol. 7 Supplement 1992 Table 1. Characteristics of Four Patients with Postarthroplasty Ankle Pain
Case
Age (years)
Sex
Onset of Ankle Pain
Duration of Ankle Pain
Index
57
F
2 weeks
9 months
2
68
F
2 weeks
6 months
3
45
M
4 weeks
Still present at 20 months
4
72
F
3 weeks
2 months
Intramedullary Pain Severity
Severe for 6 months, diminishing for 3 months Severe for 3 months, diminishing for 3 months Moderate for 6 months, still persists in mild degree Mild for 2 months
Diagnosis
Procedure
Alignment Guide
Seronegative RA
Primary cementless Miller- Galante
Yes
Loose tibial component
Revision to cemented system
Yes
Traumatic arthritis
Primary cementless Miller-Galante
Yes
Degenerative arthritis
Primary hybrid Miller- Galante
Yes, with marrow irrigation and aspiration
RA, rheumatoid arthritis.
ankle pain were first noted 6 weeks after surgery. Six months later he fell on his knee and dislocated the patellar component, requiring a patellar realignment. Twenty months after the knee arthroplasty he still complained of ankle pain. Case 4 concerned a 72-year-old woman. The intramedullary rod was used with irrigation and aspiration. Six weeks after the knee arthroplasty she complained that her ankle was more painful than her knee. Her symptoms were mild and had disappeared approximately 2 months later.
Discussion We first began using the Miller-Galante knee arthroplasty system in July 1987. From that time until November I990 we performed 87 arthroplasties. Between July I987 and December 1988, 28 arthroplasties were performed using an external tibial alignment guide. None of these patients had any complaints of ankle or shin pain. In January I989, an intramedullary tibial alignment guide became available and has been used in all cases since then. The change to the intramedullary alignment system was made because it was assumed that its use assured more accurate tibial cuts. Between January 1989 and February 1990, 30 arthroplasties were performed using the intramedullary tibial alignment rod without decompression. Three of the 30 patients developed ankle pain of sufficient severity that it interfered with their postoperative recovery. One patient still has some discomfort nearly 2 years after surgery. In February 1990, concern over possible fat emboliza-
tion 1-3 from the use of intramedullary alignment rods influenced us to begin routine irrigation and aspiration of the medullary canal prior to insertion of the alignment rod to decrease insertional pressure. Since then 29 arthroplasties have been performed, and only one patient has reported any ankle pain (case 4). Her pain was mild and lasted only 2 months. Information from t h e index case, observation of the three subsequent cases, and the notation of a decrease in incidence and severity of postoperative ankle pain once decompression was routinely used led us to postulate that this phenomenon of ankle pain results from a mechanical disturbance of the marrow contents following insertion of an intramedullary alignment guide. An effort was made to test this theory by performing a biopsy of the affected tibia, without success. It is not surprising that the biopsy did not demonstrate any abnormality, since the disturbance in the marrow is mechanical and anatomic rather than histologic.
Conclusion Because of the number of cases of fat embolism following knee arthroplasty (19 according to Motto et al.2), the use of the intramedullary alignment system has been challenged. The cases reported in this study of ankle pain following the use of an intramedullary tibial alignment system raises the additional concern of postoperative ankle pain. Experimental studies suggest that by overdrilling the proximal cortex and irrigating and aspirating the medullary canal prior to rod insertion, the insertion
Ankle Pain After TKA
pressure can be decreased and the potential complication of fat embolism reduced. ~ An additional aid in the avoidance of high insertional pressure is the use of a fluted or hollow alignment rod to which suction can be applied during insertion. We postulate that the mechanical intrusion of the m a r r o w contents into the interstices of the endosteum m a y be the cause of the postoperative ankle pain. We propose that the disturbance of the endosteal nerve supply and higher pressures aggravate the process. This theory is u n p r o v e n and is offered only for consideration.
•
Weaver and Derkash
451
References 1. Kallos T, Enis JE, Gallan F, Davis JH: Intramedullary pressure and pulmonary embolism of femoral medullary contacts in dogs during insertion of bone cement and prosthesis. J Bone Joint Surg 56A:1363, 1978 2. Monto RR, Garcia J, Callaghan J: Fatal fat embolism following total condylar knee arthroplasty. J Arthroplasty 5:291, 1990 3. Zimmerman RL, Kroner LF, Bloomberg DJ, Nallet DJ: Fatal fat embolism following total knee arthroplasty. Minn Med April:213, 1983.
Abstract: The authors present their experience with four cases of ankle pain occurring in the immediate postoperative period after total knee arthroplasty. In a 40-month period, the authors performed 87 total knee arthroplasties. In the first 28 cases, an extramedullary alignment rod was used and there were no instances of ankle pain (0%). In the next 30 cases, an intramedullary alignment rod was employed and there were three cases of moderate to severe ankle pain (10%). In the last 29 cases, the medullary canal was decompressed, irrigated, and aspirated prior to insertion of the alignment rod. In this group there was only one case of mild ankle discomfort (3.5 %). An extensive investigation was made into one of the cases in which the intramedullary alignment rod was used but the canal was not aspirated. Information from this case investigation, observation of the three subsequent cases, and the notation of a decrease in the incidence of ankle pain once aspiration was routinely used are reported. The unproven hypothesis that ankle pain may result from compression of the marrow contents into the bone interstices is presented for consideration. Three of the four cases of ankle pain reported here were spontaneously resolved within 9 months after surgery. Key words: ankle pain, intramedullary alignment, marrow disruption, intramedullary pressure, knee arthroplasty, complications, ankle pain, conservative care.
In this case we address the troublesome complication of ankle pain following total knee arthroplasty (TKA). We report our evaluation of our initial case and discuss three subsequent similar cases of ankle pain.
the right knee was performed, again using an uncemented Miller-Galante implant (Fig. 1). An intramedullary rod was used without decompression. The procedure was expeditiously accomplished without complication, and the patient was discharged 4 days after surgery on t o u c h - d o w n weight bearing using crutches. Two weeks after surgery she had her sutures removed. She progressed well with therapy (range of motion, 00-95 ° ) and had minimal knee pain, but soon began to complain of ankle and supramalleolar pain. There was no swelling or calf tenderness that would suggest thrombophlebitis. At the time of her next visit, 5 weeks after surgery, her knee had progressed well but her ankle pain had become more severe. It was intended that progressive weight bearing would begin at that time, but she was unable to bear weight because of her ankle pain. Routine radiographs of the ankle were normal (Fig. 2). At that time she was afebrile with no swelling, but the ankle and supramalleolar area remained tender.
Case Reports Index Case A 57-year-old w o m a n with seronegative rheumatoid arthritis had previous hip arthroplasty, forefoot reconstruction, contralateral left knee arthroplasty with an uncemented Miller-Galante (Zimmer, Warsaw, IN) prosthesis 7 months earlier; all procedures were without difficulty. In May 1989 arthroplasty of From the University of Colorado Health Sciences Center and Orthopaedic Associates of Aspen and Glenwood, Glenwood Springs, Colorado.
Reprint requests: James K. Weaver, MD, Orthopaedic Associates, 2001 Blake Avenue, Glenwood Springs, CO 81601.
447
448
The Journal of Arthroplasty Vol. 7 Supplement 1992
Fig. I. (A, B) Postoperative radiographs of the right knee of a 57-year-old woman with serologically negated rheumatoid arthritis (index case). A cementless first-generation Miller-Galante knee arthroplasty was performed without complication.
Ankle m o t i o n was moderately painful. There was no discoloration, hypesthesia; or temperature changes that w o u l d suggest a reflex sympathetic dystrophy. Because of concern that she m a y have developed a deep infection, blood tests were obtained. They were
normal except for a slightly elevated alkaline phosphatase. At 8 weeks after surgery, the ankle pain had not improved. The patient could not bear weight and was becoming increasingly concerned. A b o n e scan was
Fig. 2. (A) Anteroposterior and (B) lateral radiographs of the right ankle 6 weeks following TKA. At this time the patient was complaining of severe ankle pain and was unable to begin weight bearing. No abnormality is seen on the routine radiographs.
Ankle Pain After TKA
Fig. 3. Technetium bone scan of the right tibia and ankle 3 months following TKA on the ipsilateral right knee. Diffuse, intense uptake is seen throughout the entire distal portion of the tibia.
•
Weaver and Derkash
449
l
obtained to exclude the possibility of a stress fracture. The scan displayed a diffuse, intense, and increased uptake t h r o u g h o u t the distal half of the right tibia (Fig. 3). Since the bone scan did not show the focal appearance typical of a stress fracture, concern was raised about the possibility of deep sepsis or even neoplasm. At that point a magnetic resonance image was obtained (Fig. 4), which showed diffuse changes on the T-weighted image and a vaguely cylindrical outline similar to the configuration of the
intramedullary alignment rod. At 12 weeks after s u p gery, the patient continued to exhibit symptoms, was still unable to bear weight, and was very concerned about the possibility of infection or neoplasm. A Craig trocar biopsy was obtained from the medullary canal of the distal tibia, revealing n o r m a l b o n e with no inflammation or evidence of neoplasm. With this information, the patient was willing to accept our explanation that her ankle pain and related symptoms were due to the mechanical intrusion of the m a r r o w contents into the interstices of the endosteum, which was disturbing the endosteal nerve supply. Four m o n t h s after surgery she was able to begin increasing her weight bearing, and by 9 m o n t h s she was completely pain-free and had returned to her active lifestyle.
Subsequent Cases
Fig. 4. MRI T2-weighted anterioposterior projections of both tibia. Note abnormality of marrow contents on right (arrow). There is an indistinct rectangular outline similar to the cross-sectional shape of the intramedullary alignment rod.
Experience with the index case sensitized us to the possibility of ankle pain following TKA and stimulated our interest in seeking possible solutions to the problem. Three additional cases of ankle pain following TKA are identified (Table 1), t h o u g h n o n e was as severe as the index case. All four cases were analyzed with attention to the technique of implantation and were compared to the entire group. Case 2 concerned a 78-year-old w o m a n w h o h a d a knee revision for loose components. The intramedullary rod was used without decompression. Her chief postoperative complaint was ankle and shin pain. With the experience we gained f r o m the previous case, we were able to reassure her, and indeed the pain disappeared approximately 6 m o n t h s after surgery. Case 3 was a 45-year-old m a n w h o had a TKA for severe traumatic arthritis. The intramedullary rod was used without decompression. Distal tibial and
450
The Journal of Arthroplasty Vol. 7 Supplement 1992 Table 1. Characteristics of Four Patients with Postarthroplasty Ankle Pain
Case
Age (years)
Sex
Onset of Ankle Pain
Duration of Ankle Pain
Index
57
F
2 weeks
9 months
2
68
F
2 weeks
6 months
3
45
M
4 weeks
Still present at 20 months
4
72
F
3 weeks
2 months
Intramedullary Pain Severity
Severe for 6 months, diminishing for 3 months Severe for 3 months, diminishing for 3 months Moderate for 6 months, still persists in mild degree Mild for 2 months
Diagnosis
Procedure
Alignment Guide
Seronegative RA
Primary cementless Miller- Galante
Yes
Loose tibial component
Revision to cemented system
Yes
Traumatic arthritis
Primary cementless Miller-Galante
Yes
Degenerative arthritis
Primary hybrid Miller- Galante
Yes, with marrow irrigation and aspiration
RA, rheumatoid arthritis.
ankle pain were first noted 6 weeks after surgery. Six months later he fell on his knee and dislocated the patellar component, requiring a patellar realignment. Twenty months after the knee arthroplasty he still complained of ankle pain. Case 4 concerned a 72-year-old woman. The intramedullary rod was used with irrigation and aspiration. Six weeks after the knee arthroplasty she complained that her ankle was more painful than her knee. Her symptoms were mild and had disappeared approximately 2 months later.
Discussion We first began using the Miller-Galante knee arthroplasty system in July 1987. From that time until November I990 we performed 87 arthroplasties. Between July I987 and December 1988, 28 arthroplasties were performed using an external tibial alignment guide. None of these patients had any complaints of ankle or shin pain. In January I989, an intramedullary tibial alignment guide became available and has been used in all cases since then. The change to the intramedullary alignment system was made because it was assumed that its use assured more accurate tibial cuts. Between January 1989 and February 1990, 30 arthroplasties were performed using the intramedullary tibial alignment rod without decompression. Three of the 30 patients developed ankle pain of sufficient severity that it interfered with their postoperative recovery. One patient still has some discomfort nearly 2 years after surgery. In February 1990, concern over possible fat emboliza-
tion 1-3 from the use of intramedullary alignment rods influenced us to begin routine irrigation and aspiration of the medullary canal prior to insertion of the alignment rod to decrease insertional pressure. Since then 29 arthroplasties have been performed, and only one patient has reported any ankle pain (case 4). Her pain was mild and lasted only 2 months. Information from t h e index case, observation of the three subsequent cases, and the notation of a decrease in incidence and severity of postoperative ankle pain once decompression was routinely used led us to postulate that this phenomenon of ankle pain results from a mechanical disturbance of the marrow contents following insertion of an intramedullary alignment guide. An effort was made to test this theory by performing a biopsy of the affected tibia, without success. It is not surprising that the biopsy did not demonstrate any abnormality, since the disturbance in the marrow is mechanical and anatomic rather than histologic.
Conclusion Because of the number of cases of fat embolism following knee arthroplasty (19 according to Motto et al.2), the use of the intramedullary alignment system has been challenged. The cases reported in this study of ankle pain following the use of an intramedullary tibial alignment system raises the additional concern of postoperative ankle pain. Experimental studies suggest that by overdrilling the proximal cortex and irrigating and aspirating the medullary canal prior to rod insertion, the insertion
Ankle Pain After TKA
pressure can be decreased and the potential complication of fat embolism reduced. ~ An additional aid in the avoidance of high insertional pressure is the use of a fluted or hollow alignment rod to which suction can be applied during insertion. We postulate that the mechanical intrusion of the m a r r o w contents into the interstices of the endosteum m a y be the cause of the postoperative ankle pain. We propose that the disturbance of the endosteal nerve supply and higher pressures aggravate the process. This theory is u n p r o v e n and is offered only for consideration.
•
Weaver and Derkash
451
References 1. Kallos T, Enis JE, Gallan F, Davis JH: Intramedullary pressure and pulmonary embolism of femoral medullary contacts in dogs during insertion of bone cement and prosthesis. J Bone Joint Surg 56A:1363, 1978 2. Monto RR, Garcia J, Callaghan J: Fatal fat embolism following total condylar knee arthroplasty. J Arthroplasty 5:291, 1990 3. Zimmerman RL, Kroner LF, Bloomberg DJ, Nallet DJ: Fatal fat embolism following total knee arthroplasty. Minn Med April:213, 1983.
