Evaluation of patients with pathological fractures treated by standard trauma principles but neglecting the underlying malign bone disease

Serdar Demiröz, Ferhat Oktem, Aykut Çelik, Özgür Erdoğan, Korhan Özkan, Volkan Gürkan

Department of Orthopaedics and Traumatology, Kocaeli University Faculty of Medicine, ˙Izmit, Kocaeli 41001, Turkey

Department of Orthopaedics and Traumatology, Medeniyet University Faculty of Medicine, Istanbul, Turkey

Department of Orthopedics and Traumatology, Health Sciences University, Haydarpa ¸s a Numune Training and Research Hospital, Istanbul, Turkey

Department of Orthopaedics and Traumatology, Bezmialem Vakif University Faculty of Medicine, Istanbul, Turkey

ABSTRACT

Introduction: There are several studies in the literature about pathological fractures but almost no in- formation about patients whose pathological fracture caused by a malignant lesion misdiagnosed and treated as a simple fracture. The aim of this study was to investigate patient and fracture characteristics, and outcomes in cases where fractures occurred in the presence of a malign pathology but were treated as simple fractures.

Patients and Methods: Cases of malign bone lesions between 20 0 0 and 2020 were retrospectively re- viewed. Patients with a final diagnosis of malign bone lesion but whose pathological fractures were treated ignoring the underlying malign bone disease were included. Demographic, clinical and outcome data were collected from patient’s medical records and analyzed.

Results: Six patients met the inclusion criteria. Three of the patients were female and the cohort mean age was 56.8 ±21.8 years at the time of admission. Patient diagnoses were: renal cell carcinoma metas- tasis ( n = 1); colon cancer metastasis ( n = 1); chondrosarcoma ( n = 2); osteosarcoma ( n = 1); and undif- ferentiated pleomorphic sarcoma of bone ( n = 1). In all cases surgical management differed from those that should have been applied if the pathological fracture had been identified. Furthermore, surgical man- agement after definitive histological diagnosis were more aggressive compared to if the malignancy had been identified at first admission. All patients died after a mean follow-up of 16.67 ±11.7 months and the complication rate was 100%. Conclusion: When a pathological fracture is misdiagnosed and managed as a simple bone fracture, out- comes are extremely poor. In these situations, remedial surgery is more extensive, with increased com- plication rates and there is poor life expectancy.

INTRODUCTION

A pathological fracture is a fracture that occurs in a bone with reduced mechanical properties because of an underlying pathol- ogy. The pathology can be local or systemic. The various reported causes of pathological fractures include infections, benign or ma- lignant bone lesions, metabolic diseases, congenital pathologies, leukemia, and radiation therapy [1] . A pathological fracture is often discovered in an emergency room, it can usually be suspected eas- ily by careful patient history, clinical examination and radiological evaluation and finally be diagnosed by biopsy if the patient does not have a diagnosis of primary malignancy with multiple metas- tasis.

If the pathology is malign, there are usually some characteris- tic properties of both the patients and fractures. Patients may have a known malignancy, such as lung, breast or prostate or another systemic disease, the fracture mostly occurs with minor trauma (getting up from a chair or out of bed or lifting or pushing some- thing) and there might be pre-existing pain in the fractured area before the fracture occurs. A radiolucent area around the fracture site on radiograph can be seen if the lesion is lytic or there might be a sclerotic appearance if the lesion is sclerotic. The anatomi- cal region in the bone where the fracture occurs, such as the sub- trochanteric region of the femur, may also indicate a pathological fracture. However, a pathological fracture may not always be obvi- ous and may be missed. In these cases, if the underlying pathology is malignant and misdiagnosed, the result can be catastrophic. Treating a fracture without being aware of an underlying malign pathology, may result in more aggressive and complicated surg- eries, or even loss of the extremity or death.

There is extensive literature concerning pathological fractures but almost no information about misdiagnosed malignant lesions resulting in pathological fracture, which is subsequently managed as a simple fracture. The aim of this study was to retrospectively assess just such cases, including patient and fracture characteris- tics and outcomes when fractures were treated as a simple bone fracture when there was a missed underlying malignancy.

Please cite this article as: S. Demiroz, F. Oktem, A. Celik et al., Evaluation of patients with pathological fractures treated by standard trauma principles but neglecting the underlying malign bone disease, Injury, https://doi.org/10.1016/j.injury.2022.08.052

Patients and methods

After the institutional review board approval, data from cases of malign bone lesions from two different institutions, between 20 0 0 and 2020, were retrospectively reviewed. Patients whose patholog- ical fractures were only diagnosed after definitive histopathologi- cal examination and were thus initially managed as simple frac- ture were identified by examining patient histories from hospital records and included. Demographic and clinical data were collected from patient’s medical records and analyzed. Data items included age, sex, mechanism of the fracture at first admission, fractured bone, initial surgical management, histological diagnosis, presence of metastasis, surgical management following histological diagno- sis, duration between the first and second surgery, complications and duration of the follow up. The possible surgical methods ac- cording to the current literature if the fracture had been correctly diagnosed as pathological were also evaluated.

All of the patients were managed as a simple fracture at a number of hospitals but referred to the study clinics (orthopaedic oncology clinic) when suspected of having an underlying bone pathology. One of the patients had known colon cancer and mul- tiple bone metastases, so biopsy was not needed. All of the other patients had biopsy and histological diagnosis was obtained in the referral center by histopathologists who were experienced in mus- culoskeletal oncology. Whole body 99mTC-MDP skeletal scintigra- phy and thoracic computed tomography (CT) were performed for systemic evaluation.

Results

Data from 452 cases of malign bone lesions from two differ- ent institutions between 20 0 0 and 2020 were retrospectively re- viewed. Of these 432 cases, six were identified as misdiagnosed pathological fracture giving a proportion of 1.39%. Half the cases were female and the mean age was 56.8 ±21.84 years at time of admission. Patient diagnosis were: RCC metastasis ( n = 1); rec- tal adenocarcinoma metastasis ( n = 1); chondrosarcoma ( n = 2), osteosarcoma ( n = 1); and undifferentiated pleomorphic sarcoma (UPS) of bone ( n = 1). All of the lesions were in the femur and all of the patients were admitted to hospital with a pathological fracture. Two of them had occurred as a result of minor trauma and four of them occurred without any trauma ( Table 1 ). Five of the patients had no any known disease and one had an exist- ing diagnosis of rectal adenocarcinoma. He had undergone rectum resection and adjuvant chemotherapy and had been disease free for five years prior to a femur subtrochanteric fracture. Three of the patients initially underwent closed reduction and internal fix- ation with intramedullary nail (one antegrade and one retrograde femoral nail and one proximal femoral nail which had been re- moved before referral to our clinic because of the suspicion of in- fection), two of the patients had been treated by open reduction and internal fixation by proximal femoral nail and dynamic hip screw and the other patient underwent hemiarthroplasty ( Table 2 , Fig. 1 ). After the first operation, all of the patients had swelling and increasing pain. In addition, one patient had periprosthetic fracture without trauma ( Fig. 1 ). Furthermore, after the appearance of lysis rather than union at the fracture site in five and a lytic region dis- tal to the femoral stem in the patient with periprosthetic fracture, all cases were referred to the orthopaedic oncology clinic. Two of the patients had thoracic metastasis, one had bone metastasis and on had both bone and thoracic metastasis. Four of the patients had radical resection and the other two had wide resection. The mean duration between the first admission and the histological diagno- sis was 11 ±12.9 months. On review, all initial surgical manage- ment would have been different had the pathological nature of the fracture been known at the time. Furthermore, all surgical man- agement after definitive histological diagnosis was different and more aggressive from what would have been applied if the ma- lign pathology had been known at the first admission ( Table 2 ). The complication rate was 100%. Mean follow-up was 16.6 ±17.7 (range 1–42) months after second surgery, and all six patients died ( Table 3 ).

Discussion

Patients with pathological fractures are often evaluated in an emergency room, which has several disadvantages leading to diag- nostic errors, including inappropriate radiograph requests, inade- quate physical examination and patient history taking may be dif- ficult or insufficient [2–4] . A pathological fracture due to a malign lesion may occur in two situations: metastasis or primary bone malignancy. Treatment options for a patient with a pathological fracture due to malignancy differ from the options most suitable for simple bone fracture and the former should be referred to an orthopaedic oncology clinic and the treatment based on a multi- disciplinary approach. However, if a pathological fracture is misdi- agnosed and treated like a simple bone fracture, the outcome can be catastrophic. Fracture with minor or even no trauma and pre-existing pain at the fracture site strongly suggest a pathological fracture and these might be the most important clues to establish diagnosis. Hu et al. reported that over 75% of fractures due to bone metastases occurred in the absence of trauma. They also reported that pain was present in 85% of patients before the fracture occurred [3] . In a study by O’Flaherty et al. in patients with spontaneous femoral neck fractures and known malignancies, the histopathological find- ings consistently indicated bone metastases, even when the imag- ing studies failed to show any signs of malignancy [5] . Two of the patients in the current study had a fracture because of a simple fall and four patients had no identifiable trauma. Four of the pa- tients had pre-existing pain at the fracture site although data was incomplete on the other two patients. In the patient with a known history of colon cancer there was a radiograph before the fracture occurred, performed because of the pain, but metastasis was not suspected ( Fig. 2 ). In this series, having pre-existing pain and mi- nor or absent trauma were the most important and consistent find- ings suggesting a pathological fracture. Radiologic appearance and properties of the fracture site are also important to assist in the diagnosis of pathological fracture. If the quality of radiographs is poor, good-quality antero-posterior and lateral views must be obtained before treatment. Characteris- tics seen in bone malignancies can include marked lysis or a ma- trix calcification or ossification, cortical destruction, a periosteal re- action and spread to adjacent soft tissues [6] . Furthermore, most of the bone metastases occur in the proximal femur [7] . Avulsion fracture of the lesser trochanter and subtrochanteric transverse fractures are also suggestive of bone metastasis. Hu et al. reported that clinical evaluation combined with analysis of the radiographs established the diagnosis of pathological fracture in more than 85% of cases [3] . All of the fractures in the current study were in the fe-

Fig. 1. Radiographs of the patients after referral to the orthopaedic oncology clinic because of increasing pain, swelling and nonunion, periprosthetic fracture or extensive bone destruction on x-ray at follow up. ( a) Extensive lysis around the proximal femoral nail in Patient 1; ( b) Patient 2 was referred after removal of a proximal femoral nail because of suspicion of infection due to pain and lytic appearance; ( c) extensive lysis around stem and further lytic areas distal to the stem resulted in periprosthetic fracture in Patient 3; ( d) marked lysis around trochanter minor in Patient 4; ( e) marked lysis around the fracture instead of union in Patient 5; and ( f) again marked lysis around the fracture rather than union in Patient 6.

mur. Unfortunately, we could only review three patient radiographs at the time the fracture occurred but all were typically lytic with cortical thinning ( Fig. 3 ). There are several published studies about misdiagnosed patho- logical fractures caused by benign pathology, but studies of mis- diagnosed malign pathological fractures are rare [ 1 , 8 , 9 ]. Kong et al reported a case in which a breast cancer metastasis caused femoral neck fracture [10] . The patient was a 38-year-old woman with known breast cancer. During her cancer treatment she experienced left hip pain. She underwent radiographs and positron emission to- mography scans but the pathological fracture was missed. The pa- tient was instructed to continue chemotherapy. Almost two years after the start of the hip pain, the patient was evaluated by an orthopaedic oncologist, and femoral neck fracture was detected

Fig. 2. (a) X-rays of Patient 1 before pathological fracture occurred performed because of pain. Cortical thinning and lysis can be seen; (b) X-ray of pathological fracture in the subtrochanteric region; (c) extensive bone destruction on X-ray at follow up after open reduction and internal fixation; and (d) after hip disarticulation.

Fig. 3. Patient’s radiographs at the time of fracture, all typically lytic with cortical thinning. ( a) Lytic area in subtrochanteric region in Patient 1; ( b) lytic area in in- tertrochanteric region in Patient 4; and ( c) Cortical thinning in the posterior aspect of the proximal fragment at the fracture line in Patient 6.

which had healed spontaneously and the patient was pain free. Heaton et al. reported a missed pathological fracture from multi- ple myeloma [11] . A 68-year-old man presented with pain and ec- chymosis in the third finger after having a fall. Radiographs con- firmed non-displaced proximal phalanx fracture and the patient was treated with an orthosis for one month. After one month, he presented with persistent pain and the radiographs showed no healing with a lytic lesion adjacent to the fracture. A subsequent examination of the patient’s medical history revealed that he had been in remission for 12 years following treatment for multiple myeloma. The tumor was removed and cemented joint arthroplasty was performed. The patient had no pain with 0–70 °joint range of motion three months postoperatively. In these cases, femoral neck fracture was not diagnosed in the former, and the chosen treatment was initially conservative in the latter. Therefore, none of the patients in these two case reports had undergone in-

Fig. 4. Radiographs of the patients treated with radical resection and megaprosthesis as a result of dissemination of the tumor into the whole compartment. ( a) Radiographs of fracture treated by Dynamic Hip Screw (DHS), radical resected total femur and reconstruction with total femur prosthesis in Patient 2; ( b) fracture treated by retrograde nail, resected distal three-quarters of the femur with nail and reconstruction with distal femoral resection prosthesis in Patient 3; ( c) fracture treated by antegrade nail, resected total femur and reconstruction with total femur prosthesis in Patient 4; ( d) fracture treated by proximal femoral nail and then removed and reconstructed with proximal femur resection prosthesis in Patient 5; and ( e) Fracture treated by hemiarthroplasty and then reconstructed with total femur prosthesis and ice cone acetabular prosthesis in Patient 6.

appropriate surgical intervention, which would lead to more ex- tensive surgery later. Herrera-Pérez et al. reported an 86 year-old woman who presented with swelling and pain, three months after surgical treatment of humeral diaphysis with intramedullary pin- ning [12] . Radiograph showed complete destruction of almost all the humerus. Open biopsy was performed and the diagnosis was low-grade myofibroblastic sarcoma. The patient past medical his- tory also revealed that she was treated for myofibroblastic sarcoma of the maxilla a year earlier. Due to the age of the patient and end stage of disease (multiple metastases) the clinical decision was to give palliative treatment only. Otherwise, amputation would have been the most probable treatment option because of complete de- struction of the humerus and soft tissue enlargement.

Jeon et al. evaluated 39 patients with a diagnosis of osteosar- coma over 40 years old [13] . Eleven cases were misdiagnosed, of which three had been initially diagnosed with simple fracture. These authors indicated that these three patients had been treated with routine conventional protocols but there was no data con- cerning surgical management or outcome. Sadoghi et al. evaluated patients with a diagnosis of osteosarcoma over 60 years old [14] . Three of the seven patients included had been misdiagnosed. Two of three was misdiagnosed as fibrosis dysplasia and juvenile cyst. One patient was misdiagnosed as a simple fracture in the diaph- ysis of the femur and was treated by intramedullary nailing. Then, because of the dissemination into the whole femoral medulla, total femoral resection was needed. To the best of our knowledge, this is the only report in which the patient was treated with another extensive surgical procedure as a result of misdiagnosis of malign pathologic fracture and treatment as a simple bone fracture. Again, the authors did not give a detailed evaluation of the patient’s treat- ment and follow-up. Both of these studies emphasized the higher risk of misdiagnosis of osteosarcoma in older ages due to atypical radiological findings. In contrast, in the current study there was only one patient with osteosarcoma. Although osteosarcoma is the most common primary malign bone tumor, it is mostly diagnosed between the ages of 10 and 30. It has a second peak in patients older than 65 but chondrosarcoma is the most common primary bone sarcoma in adults and should be kept in mind if pathological fracture is suspected.

Metastases are the most common malign tumors of bone and they are much more common than primary bone tumors, so most of the pathologic fractures related to malign bone lesion are be- cause of metastasis [ 15 , 16 ]. In contrast, in the current study there were only two patients with metastatic lesions. Although a patho- logic fracture might be a first sign of a cancer, in most of the cases it indicates an end stage disease and if the patient has a known primary malignancy, to establish a diagnosis of pathologic fracture would be easier. Besides, patients with metastatic bone lesions are mostly in the last stage of their life and the surgical treatment is mostly palliative. Then, lower survival rates can explain the de- creased rate (33.3%) of patients with metastatic lesions in this co- hort, in which patients die before the treatment related complica- tions occur. However, if there is no history of malignancy or if the patient is young, the diagnosis of metastasis may be challenging but there are also several clues to indicate metastatic disease. Es- pecially if the patient is older than 50 and has had a long bone fracture, the consultation should include questions about general health status. A decline in general health and recent weight loss may indicate a metastatic disease. In the current study, one of the two patients with metastatic disease was an 83-year-old woman with no known malignancy and the other was a 32-year-old man who has a history of rectal adenocarcinoma but had been disease- free for five years. If there is suspicion of pathological fracture as a result of clinical and radiological assessment, a metastatic lesion should be kept in mind and patients should be asked about symp- toms and other factors associated with specific types of primary cancer, such as recurrent respiratory tract infections, current smok- ing, chronic obstructive lung disease, alcoholism, and cirrhosis of the liver, and the medical history of the patients should be care- fully reviewed, even in young patients.

When a malignant pathological fracture is misdiagnosed and managed as a simple bone fracture, the patient is likely to need further extensive surgery with high complication rates and, un- fortunately, efforts to save the limb or the patient life are most likely unsuccessful. In the current study all of the patients had fractures in the diaphysis or proximal femur. The preferred meth- ods for surgical treatment of simple fracture in these regions are intramedullary nailing in the diaphysis, hemiarthroplasty, proximal femoral nailing, anatomical plate or DHS in the proximal femur. All of the methods lead to dissemination and contamination of the whole compartment while reaming or scraping for intramedullary nailing or hemiarthroplasty. Plate fixation also leads to contami- nation during open reduction of the fracture ( Fig. 4 ). As a result, three of the patients had total femoral resection arthroplasty, one had distal femoral resection arthroplasty, one had proximal femoral resection arthroplasty with type 2 hemipelvectomy and one had hip disarticulation. All of the patients had complications, such as infection, surgical site necrosis, and recurrence with a mean sur- vival of only 16.67 months after the last surgery ( Table 3 ). Having a history of previous surgery, a weaker immune system because of the malignancy, delayed diagnosis and extensive surgical approach for radical and wide resections may explain the high complication rate in these patients.

In conclusion, we reported a series of six patients who needed further extensive and risky surgical intervention with high compli- cation rates because of misdiagnosed pathological fracture, treated as a simple fracture, which is the largest series in the literature to date. Fractures in femur with no or only minor trauma and pre- existing pain at the fracture site should be cause for alarm. Radio- graphs may contain clues to the correct diagnosis and careful eval- uation of these is essential to establish a diagnosis of pathological fracture. However, in our opinion, if the possibility of pathological fracture is kept in mind, it is easy to diagnose. This is important because if the diagnosis is missed, what can be done is limited and the results are extremely poor.

Funding declaration

No funding received for this work from any agency.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Com- mittee of Kocaeli University.

Consent to participate

Comprehensive agreement for academic use of information such as type of treatments and treatment progress and any data acquired during their treatments was also obtained from the pa- tients by the hospital at the time of their hospitalization.

Consent to publish

There is no any individual person’s data.

Availability of data and materials

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Declaration of Competing Interest

The authors have no relevant financial or non-financial interests to disclose.

CRediT authorship contribution statement

Serdar Demiroz: Visualization, Writing –original draft, Data curation, Formal analysis. Ferhat Oktem: Visualization, Writing –original draft, Data curation, Formal analysis. Aykut Celik: Visu- alization, Writing –original draft, Data curation, Formal analy- sis. Ozgur Erdogan: Visualization, Writing –original draft. Korhan Ozkan: Visualization, Writing –original draft, Writing –review & editing. Volkan Gurkan: Visualization, Writing –original draft, Writing –review & editing.

Acknowledgments

None.

References

Canavese F, Samba A, Rousset M. Pathological fractures in children: diagnosis and treatment options. Orthop Traumatol Surg Res 2016;102:149-59. doi:10. 1016/j.otsr.2015.05.010.

Cheung FH. The practicing orthopedic surgeon’s guide to managing long bone metastases. Orthop Clin North Am 2014;45:109-19. doi:10.1016/j.ocl.2013.09. 003.

Hu YC, Lun DX, Wang H. Clinical features of neoplastic pathological fracture in long bones. Chin Med J (Engl) 2012;125:3127-32.

Wedin R. Surgical treatment for pathologic fracture. Acta Orthop Scand 2001;72:2p.

O’Flaherty MT, Thompson NW, Ellis PK, Barr RJ. Full-length radiographs of the femur in patients with a femoral neck fracture and co-existent malignancy-are they benefit? Ulst Med J 2008;77:181-4.

Janssen SJ, van der Heijden AS, van Dijke M, Ready JE, Raskin KA, Ferrone ML, et al. Marshall Urist Young Investigator Award: prognostication in patients with long bone metastases: does a boosting algorithm improve survival estimates? Clin Orthop Relat Res 2015;473:3112-21. doi:10.1007/s11999-015-4446-z.

Jacofsky DJ, Haidukewych GJ. Management of pathologic fractures of the prox- imal femur: state of the art. J Orthop Trauma 2004;18:459-69. doi:10.1097/ 00005131-200408000-00013.

Sferopoulos NK. Subtrochanteric osteoid osteoma: a misdiagnosed case com- plicated by a hip fracture. Chin J Traumatol 2016;19:283-5. doi:10.1016/j.cjtee. 2016.03.006.

Hoeffel C, Panuel M, Plenat F, Mainard L, Hoeffel JC. Pathological fracture in non-ossifying fibroma with histological features simulating aneurysmal bone cyst. Eur Radiol. 1999;9:669-71. doi:10.1007/s003300050730.10.

Kong AC, Zarate SD, Belzarena AC. Missed pathological femoral neck fracture undergoes spontaneous healing. Radiol Case Rep 2021;17:72-6. doi:10.1016/j. radcr.2021.10.002.11.

Heaton D, Alexander H, Trumble TE. Missed pathologic fracture from multiple myeloma. J Hand Surg Am 2015;40:1501-3. doi:10.1016/j.jhsa.2014.07.037.12.

Herrera-Perez M, Boluda-Mengod J, Munoz-Ortus R, Gutierrez-Morales MJ, Pais-Brito J. Continuous pain and swelling after humerus fracture in an 86-years-old woman. Acta Ortop Mex 2017;31:30-4.

Jeon DG, Lee SY, Cho WH, Song WS, Park JH. Primary osteosarcoma in patients older than 40 years of age. J Korean Med Sci 2006;21:715-18. doi:10.3346/ jkms.2006.21.4.715.14.

Sadoghi P, Leithner A, Clar H, Glehr M, Wibmer C, Bodo K, Quehenberger F, Windhager R. The threat of misdiagnosis of primary osteosarcoma over the age of 60: a series of seven cases and review of the literature. Arch Orthop Trauma Surg 2010;130:1251-6. doi:10.1007/s00402-009-1011-9.15.

Rougraff BT, Kneisel JS, Simon MA. Skeletal metastases of unknown ori- gin: a prospective study of a diagnostic strategy. J Bone Joint Surg Am 1993;75:1276-81.

Mankin HJ, Lange TA, Spanier SS. The hazards of biopsy in patients with malignant primary bone and soft tissue tumors. J Bone Joint Surg Am 1982;64:1121-7.