Abstract
We performed a retrospective study of thirteen patients who had had sixteen pathological fractures of the shaft of the humerus secondary to metastatic disease. All but one fracture was stabilized with interlocking intramedullary nailing with use of a closed technique.The mean operative time for the sixteen procedures was ninety-two minutes (range, fifty to 180 minutes), the mean blood loss was 116 milliliters (range, fifty to 200 milliliters), and the mean duration of hospitalization was five days (range, two to ten days). Fourteen extremities had a return to nearly normal function within three weeks after nailing. Relief of pain was rated as good or excellent in all but one patient. Eleven patients (fourteen humeri) received radiation therapy at a mean of seven days (range, three to fourteen days) after the operation. Nine patients died at a mean of four months (range, one to twelve months) postoperatively; the remaining four patients were still alive at a mean of ten months (range, nine to fifteen months). There were no problems related to the wound, deep infections, nerve palsies, or failures of the implant. The fracture was united in all seven of the eleven extremities in patients who survived for at least three months and had radiographs available.Interlocking intramedullary nailing of the humerus for pathological fractures provides immediate stability and can be accomplished with a closed technique, brief operative time, and minimum morbidity, with a resultant early return of function to the extremity.
Pathological fractures that are due to metastatic tumors and related lesions are commonly seen in orthopaedic practice. In adults, metastatic disease is the most common malignant neoplasm of bone1,7,12. More than 150,000 patients are newly diagnosed as having bone metastasis each year18. The femur is the most frequently involved long bone. The humerus is the next most commonly affected long bone, accounting for 16 per cent (ninety-six) of 588 pathological fractures in long bones in one series7.
Pathological fractures of the humerus usually occur late in the course of metastatic disease6. Severe pain accompanies the injury, and most patients lose functional use of the arm. Non-operative treatment rarely leads to adequate relief of pain or to a return of function. Radiation therapy alone provides some relief of pain but may further delay fracture-healing and the return of function8,13. The loss of the use of any limb increases the requirements for nursing care and institutionalization. This reduces the quality of the patient's life and increases society's expense for the care of the patient.
Poor results from the non-operative treatment of pathological fractures of long bones have led many to recommend operative stabilization5-7,13. Operative fixation of pathological fractures of the humerus may involve intramedullary fixation with use of smooth pins, often with an open debulking of the tumor and packing with polymethylmethacrylate. However, interlocking intramedullary nailing systems that allow closed reduction techniques are now available for the treatment of these difficult lesions and may provide some benefit compared with other devices.
To our knowledge, there have been no previous reports of the results of treatment of pathological fractures of the humerus with these locking intramedullary devices. The purpose of this retrospective study was to analyze the results of closed interlocking intramedullary fixation of pathological fractures of the humerus.
*One or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Section of Orthopaedic Surgery, University Hospitals, University of Michigan, 1500 East Medical Center Drive, Ann Arbor, Michigan 48109-0328.
‡Department of Orthopaedic Surgery, Grace Hospital, 6071 West Outer Drive, Detroit, Michigan 48235.
Thirteen patients who had fifteen pathological fractures and one impending fracture of the humerus were managed with interlocking intramedullary nailing at the University of Michigan Hospitals or the affiliated Veterans Hospital in Ann Arbor between 1990 and 1994. The charts and radiographs of all patients were available for retrospective review. All fractures were of the humeral diaphysis and were considered to be appropriate for nailing, as they were between the proximal one-sixth and the distal one-fourth of the humerus. (Fractures outside of this region were given an alternative form of treatment.) All patients who had operative treatment for metastatic disease of the humeral diaphysis were included in the study. A painful pathological fracture or impending fracture was the primary indication for the operation. The patient who had an impending fracture had a large, painful, lytic lesion and a history of a pathological fracture of the contralateral humerus. Other prerequisites for inclusion in the study were an expected survival of more than six weeks and sufficient medical fitness to tolerate an anesthetic. Intramedullary nailing was offered only to patients who met these criteria.
Of the thirteen patients, six were women and seven were men (Table I). The mean age of the patients was sixty-three years (range, forty-four to seventy-five years). The diagnoses included multiple myeloma in four patients, breast cancer in two, prostate cancer in two, and lung cancer, renal-cell carcinoma, lymphoma, melanoma, and an adenocarcinoma of unknown origin in one patient each. Four patients had had additional femoral lesions necessitating operative treatment; two of these procedures had been performed before the index operation and two, after it. The mean interval between the diagnosis of the malignant tumor and the pathological fracture was fifty-two months (range, zero to 144 months). In two patients, the malignant tumor was diagnosed after the pathological fracture of the humerus. All patients had moderate or severe pain with limited function before the fixation.
Operative Technique
The Uniflex humeral nail (Biomet, Warsaw, Indiana) was used in all sixteen extremities. This nail can be locked with as many as four proximal and two distal screws.
All but one procedure was performed with a closed technique, to decrease disruption of the soft tissues and to improve the mobility of the limb postoperatively. Early postoperative irradiation is possible when a closed technique is used. One patient (Case 6) had an open procedure because an open biopsy was necessary to retrieve tissue for histological confirmation of recurrence of lymphoma, and the surgeon took advantage of the already opened fracture site to facilitate reduction. All nails were inserted antegrade through the proximal aspect of the humerus.
The lateral decubitus position allowed for the most convenient use of fluoroscopy and was employed for fourteen fractures (twelve patients). A supine or beach-chair position was used for one patient who had bilateral fracture. A deltoid-splitting incision was started at the anterior aspect of the acromioclavicular joint and extended about 2.5 centimeters. The site of entry for the nail into the humerus was in the sulcus between the greater tuberosity and the articular surface of the humeral head, directly in line with the medullary canal. The supraspinatus tendon was split in line with its fibers to expose this entry site. A guide-wire was passed across the fracture site with fluoroscopic guidance. The canal was then reamed to one millimeter greater than the size of the nail. One nail had a diameter of eight millimeters and fifteen had a diameter of nine millimeters. The lengths of the nails ranged from twenty-two to twenty-eight centimeters. The nail was inserted over the guide-wire. Proximal interlocking was performed with use of 4.5-millimeter screws with the assistance of a short nail-mounted drill-guide. Fourteen fractures were fixed with two proximal locking screws and two, with three proximal locking screws. Distal interlocking was performed with the Uniflex nail system with the assistance of a long nail-mounted drill-guide. The relatively short, straight span of the nail and the minimum deforming forces in the humerus permitted accurate distal interlocking. For placement of the distal locking screws, a limited dissection was carried out through a five-centimeter incision down to bone to ensure that the radial nerve was protected, and the 4.5-millimeter screws were inserted from the lateral side. Fifteen fractures were fixed with two distal locking screws and one was fixed with one distal locking screw.
The mean operative time for the sixteen procedures was ninety-two minutes (range, fifty to 180 minutes). The mean blood loss was 116 milliliters (range, fifty to 200 milliliters). The mean duration of hospitalization was five days (range, two to ten days). The mean interval between the pathological fracture (or the discovery of an impending fracture) and the operative treatment was four days (range, zero to thirteen days). Many patients were referred from other institutions to facilitate the treatment of the primary tumor, and coexisting medical problems sometimes delayed fixation of the fracture. One patient (Case 12), who had severe pain, was considered too ill, because of poor cardiac function, to tolerate a general anesthetic. He had intramedullary nailing with a local anesthetic and survived for four months postoperatively.
Postoperative Regimen and Follow-up
Use of the arm was encouraged immediately after the operation. No external supports were used. Patients were instructed with regard to exercises and were allowed to use assistive devices for walking if needed.
All patients received chemotherapy or radiation therapy, or both, as adjuvant treatment. Four patients had received chemotherapy and three, radiation therapy before they sustained the humeral fracture. Postoperatively, four patients received chemotherapy. Eleven patients (fourteen humeri) received radiation therapy postoperatively, with treatments begun a mean of seven days after fixation.
All patients were seen postoperatively by one of the senior ones of us (J. S. B. or R. B. B.). The follow-up examination included evaluation of the range of motion of the shoulder and the functional use of the extremity. Preoperative and postoperative pain was quantified according to the scale described by Perez et al. With use of this scale, relief of pain is assessed on the basis of the patient's use of analgesics. An excellent result is defined as complete resolution of pain, with no use of analgesics; a good result, as nearly complete resolution of pain, with occasional use of analgesics; a fair result, as some decrease in pain but regular use of analgesics; and a poor result, as no difference in pain compared with the level before treatment13. For patients who had chronic pain, the scale was interpreted to reflect the increased amount of narcotics used acutely as a result of the humeral fracture as compared with the previous, baseline requirements.
There was good relief of pain within two weeks after ten nailing procedures and excellent relief after five (Figs. 1-A and 1-B). One patient (Case 10), who had advanced metastatic melanoma and a concurrent pathological fracture of the femur, had only fair relief of pain, with persistent discomfort in the arm. She had fallen at the time of the humeral fracture and had also sustained multiple fractures of the ribs. Another patient (Case 3), after a two-month interval of good function and relief of pain, had a late local recurrence of a pulmonary metastasis associated with return of pain but no loss of fixation. He died one week after the recurrence was noted. He had received a conventional (2000-centigray) postoperative dose of radiation for control of the tumor, but this treatment clearly had failed locally.
All patients had improvement in the functional use of the extremity after fixation. No patient noted any limitation of motion of the elbow. The range of motion of the shoulder was documented numerically for ten patients, who had a mean of 101 degrees (range, 55 to 180 degrees) of abduction and 98 degrees (range, 45 to 170 degrees) of forward flexion. Two other patients had a full range of motion of the shoulder. Eleven of the thirteen patients recovered full use of the limb for activities of daily living and for walking with use of assistive devices at a mean of two weeks postoperatively. Two patients (Cases 8 and 10) had had weakness preoperatively that was due to metastatic involvement of the cervical spine. Both of these patients had an improvement in the functional use of the extremity but neither was fully independent with regard to activities of daily living because of continued weakness. There was evidence of osseous union in all seven of the eleven extremities in patients who survived for at least three months and had radiographs available.
Complications were infrequent. There were no infections, problems related to the wound, instances of postoperative drainage, or failures of fixation or of the implant. One patient (Case 1) had removal of a proximal locking screw because of local tenderness.
Nine patients died, at a mean of four months (range, one to twelve months) after fixation of the pathological fracture. Two patients were alive one year postoperatively. One patient lived for only one month after fixation. The mean duration of follow-up for the four surviving patients was ten months (range, nine to fifteen months).
Closed treatment of pathological fractures of the humerus has been relatively unsuccessful. Douglass et al. reported on a series of patients who were managed non-operatively. Nine of these patients had had radiation therapy, and only five of them had relief of pain; four had a persistently functionless limb. Of twelve patients who had been managed with immobilization alone, only three had relief of pain. Flemming and Beals reported similarly discouraging results, with three of eight patients having poor relief of pain and four, poor function. Perez et al. reported on twenty-one patients who had been managed non-operatively for a pathological fracture of the upper limb; nine had fair or poor function and seven had poor relief of pain. In contrast, of nine patients who had been managed operatively, all had good or excellent function and relief of pain13.
In a large, multi-institutional study, Harrington et al. demonstrated benefit from internal fixation and augmentation with polymethylmethacrylate for the treatment of pathological fractures that were due to metastasis. Their study included sixty-eight humeral fractures, most of which were treated with Rush rods with or without cement, with a good result.
Intercalary segmental replacement with a metal spacer for the treatment of pathological fractures of the humeral diaphysis was described by Chin et al., in a series of four patients from the Mayo Clinic. This technique requires an extensive exposure of the fracture for excision of the involved bone. Biomechanical testing showed the segmental device to be a relatively stronger construct compared with fixation of the fracture, and the results were excellent. However, use of this technique has not become widespread.
Lewallen et al. reported on fifty-five patients in whom a pathological fracture of the humerus had been treated with intramedullary nailing with use of Rush rods at the Mayo Clinic. Most of the rods were inserted with augmentation with polymethylmethacrylate, but some were not. All but one patient had improved function. The mean duration of survival after treatment was ten months, and the mean operative time was eighty or 110 minutes depending on whether cement was used. Seven of the fifty-five patients had recurrent pain, and three revision operations were performed11. Yazawa et al. reported on a later series, from the same institution, that included both femoral and humeral metastatic lesions. Of thirty-five operatively treated humeral lesions, thirty were treated with intramedullary nailing and most had augmentation with polymethylmethacrylate. The mean operative time was 118 minutes, and the mean estimated intraoperative blood loss was 437 milliliters18.
Intramedullary fixation offers an advantage compared with plate fixation in that the implant is load-sharing rather than load-bearing. The locking intramedullary nails used in the current study differed mechanically from the Rush rods used by Lewallen et al. The larger diameter of the nail as well as the locking screws allows for rotational and axial control. Although our series was small, the mean operative time was comparable with that reported by Lewallen et al., and the rate of reoperation was lower. We had no failures of fixation, even though cement had not been used. This finding is most likely related to the interlocking capability and the larger diameter of the nails that we employed.
Lancaster et al. reported their experience with the treatment of pathological fractures of the humerus and, like other investigators, described poor results after non-operative treatment. They noted complications in association with seven of forty-six fractures that had been treated operatively with use of various devices. The best results were in patients who had been managed with Küntscher nails; twenty-six of twenty-nine had good or excellent relief of pain, and twenty-two of twenty-three for whom information was available had good functional recovery. The procedures in that series were nearly always performed open and required augmentation with polymethylmethacrylate. In our series, the need for cement was obviated by the fixation provided by the locking screws.
In the current series, stable fixation was achieved in almost all patients with use of a closed technique. The only open reduction was associated with an open biopsy that was needed for diagnosis. Use of a closed technique allowed early postoperative irradiation. Radiation therapy was begun by a mean of seven days after the operation, with some patients starting the therapy before they were discharged from the hospital.
Although this technique is effective for the treatment of diaphyseal lesions, interlocking intramedullary nailing should not be considered for fractures outside of the area of interlocking, which includes the distal one-fourth and the proximal one-sixth of the humerus. Fractures of the proximal aspect of the humerus may best be treated with hemiarthroplasty. Options for the treatment of pathological fractures of the distal one-fourth of the humerus include application of a plate and resection of the distal part of the humerus, with prosthetic reconstruction of the elbow4,15.
The decision to proceed with operative treatment of metastatic disease must be carefully considered in light of the patient's functional status before the fracture, life expectancy, type of tumor, and anatomical extent of local and systemic involvement. Pathological fractures of the shaft of the humerus often occur late in the course of the disease. Operative intervention rarely should be considered in patients who have an anticipated life expectancy of less than six weeks. However, judicious operative intervention can improve the quality of the terminal portion of a patient's life3,8,9,13,14,16,17. Although we did not specifically address this issue in the current study, it seems likely that the cost of operative treatment of these fractures in appropriately selected patients is partially or completely offset by the increased ability of the patient to use the upper limb for hygiene and feeding, thus reducing the amount of supportive and nursing care that is needed. If the value to the patient of the improved quality of life is considered, then the analysis of operative treatment of pathological fractures of the humerus is even more likely to indicate benefit from such treatment.
NOTE: The authors acknowledge Thomas J. Graham, M.D., Orthopaedic Department, Cleveland Clinic, who gathered a portion of the data used in this study.
Brage, M. E., and |and |Simon, M. A.: Evaluation, prognosis, and medical treatment considerations of metastatic bone tumors. Orthopedics,15: 589-596, 1992.15589
1992
[PubMed]
Chin, H.-C.; Frassica, F. J.; Hein, T. J.; Shives, T. C.; Pritchard, D. J.; Sim, F. H.; and |and |Chao, E. Y. S.: Metastatic diaphyseal fractures of the shaft of the humerus. The structural strength evaluation of a new method of treatment with a segmental defect prosthesis. Clin. Orthop.,248: 231-239, 1989.248231
1989
[PubMed]
Colyer, R. A.: Surgical stabilization of pathological neoplastic fractures. Curr. Prob. Cancer,10: 117-168, 1986.10117
1986
[CrossRef]
Damron, T. A.; Heiner, J. P.; Freund, E. M.; Damron, L. A.; McCabe, R.; and |and |Vanderby, R.: A biomechanical analysis of prophylactic fixation for pathological fractures of the distal third of the humerus. J. Bone and Joint Surg.,76-A: 839-847, June 1994.76-A839
1994
Douglass, H. O., Jr.; Shukla, S. K.; and |and |Mindell, E.: Treatment of pathological fractures of long bones excluding those due to breast cancer. J. Bone and Joint Surg.,58-A: 1055-1061, Dec. 1976.58-A1055
1976
Flemming, J. E., and |and |Beals, R. K.: Pathologic fracture of the humerus. Clin. Orthop.,203: 258-260, 1986.203258
1986
[PubMed]
Habermann, E. T., and |and |Lopez, R. A.: Metastatic disease of bone and treatment of pathological fractures. Orthop. Clin. North America,20: 469-486, 1989.20469
1989
Harrington, K. D.; Sim, F. H.; Enis, J. E.; Johnston, J. O.; Dick, H. M.; and |and |Gristina, A. G.: Methylmethacrylate as an adjunct in internal fixation of pathological fractures. Experience with three hundred and seventy-five cases. J. Bone and Joint Surg.,58-A: 1047-1055, Dec. 1976.58-A1047
1976
Jensen, T. M.; Dillon, W. L.; and |and |Reckling, F. W.: Changing concepts in the management of pathological and impending pathological fractures. J. Trauma,16: 496-502, 1976.16496
1976
[PubMed][CrossRef]
Lancaster, J. M.; Koman, L. A.; Gristina, A. G.; Rovere, G. D.; Poehling, G. G.; Nicastro, J. F.; and |and |Adair, D. M.: Pathologic fractures of the humerus. Southern Med. J.,81: 52-55, 1988.8152
1988
[PubMed][CrossRef]
Lewallen, R. P.; Pritchard, D. J.; and |and |Sim, F. H.: Treatment of pathologic fractures or impending fractures of the humerus with Rush rods and methylmethacrylate. Experience with 55 cases in 54 patients, 1968-1977. Clin. Orthop.,166: 193-198, 1982.166193
1982
[PubMed]
Miller, F., and |and |Whitehill, R.: Carcinoma of the breast metastatic to the skeleton. Clin. Orthop.,184: 121-127, 1984.184121
1984
[PubMed]
Perez, C. A.; Bradfield, J. S.; and |and |Morgan, H. C.: Management of pathologic fractures. Cancer,29: 684-693, 1972.29684
1972
[PubMed][CrossRef]
Pritchard, D. J.: Lesions of the humerus. In Diagnosis and Management of Metastatic Bone Disease: A Multidisciplinary Approach, pp. 213-220. Edited by F. H. Sim. New York, Raven Press, 1988.
Pritchard, D. J.: Pathologic fractures of the humerus. Orthopedics,15: 557-562, 1992.15557
1992
[PubMed]
Sim, F. H., and |and |Pritchard, D. J.: Metastatic disease in the upper extremity. Clin. Orthop.,169: 83-94, 1982.16983
1982
[PubMed]
Vail, T. P., and |and |Harrelson, J. M.: Treatment of pathologic fractures of the humerus. Clin. Orthop.,268: 197-202, 1991.268197
1991
[PubMed]
Yazawa, Y.; Frassica, F. J.; Chao, E. Y. S.; Pritchard, D. J.; Sim, F. H.; and |and |Shives, T. C.: Metastatic bone disease. A study of the surgical treatment of 166 pathologic humeral and femoral fractures.. Clin. Orthop.,251: 213-219, 1990.251213
1990
[PubMed]