Abstract
Aims
The modern prevalence of primary tumours causing metastatic bone disease is ill-defined in the oncological literature. Therefore, the purpose of this study is to identify the prevalence of primary tumours in the setting of metastatic bone disease, as well as reported rates of pathological fracture, postoperative complications, 90-day mortality, and 360-day mortality for each primary tumour subtype.
Methods
The Premier Healthcare Database was queried to identify all patients who were diagnosed with metastatic bone disease from January 2015 to December 2020. The prevalence of all primary tumour subtypes was tabulated. Rates of long bone pathological fracture, 90-day mortality, and 360-day mortality following surgical treatment of pathological fracture were assessed for each primary tumour subtype. Patient characteristics and postoperative outcomes were analyzed based upon whether patients had impending fractures treated prophylactically versus treated completed fractures.
Results
In total, 407,893 unique patients with metastatic bone disease were identified. Of the 14 primary tumours assessed, metastatic bone disease most frequently originated from lung (24.8%), prostatic (19.4%), breast (19.3%), gastrointestinal (9.4%), and urological (6.5%) malignancies. The top five malignant tumours resulting in long bone pathological fracture were renal (5.8%), myeloma (3.4%), female reproductive (3.2%), lung (2.8%), and breast (2.7%). Following treatment of pathological fractures of long bones, 90-day mortality rates were greatest for lung (12.1%), central nervous system (10.5%), lymphoma (10.4%), gastrointestinal (10.1%), and non-renal urinary (10.0%) malignancies. Finally, our study demonstrates improved 90-day and 360-day survival in patients treated for impending pathological fracture compared to completed fracture, as well as significantly lower rates of deep vein thrombosis, pulmonary embolism, urinary tract infection, and blood transfusion.
Conclusion
This study defines the contemporary characteristics of primary malignancies resulting in metastatic bone disease. These data should be considered by surgeons when prognosticating patient outcomes during treatment of their metastatic bone disease.
Cite this article: Bone Jt Open 2023;4(6):424–431.
Take home message
This is the largest modern cohort of patients with metastatic disease to bone.
With increases in survival and treatment options, the primary sites of disease are shifting.
Patients who are surgically stabilized for impending pathological fractures have lower complication rates and improved survival, compared to patients treated for completed pathological fractures.
Introduction
Despite advances in medical and surgical oncology, malignancy remains one of the leading causes of mortality worldwide.1 In 2020 alone, there were approximately ten million deaths worldwide due to cancer.1 One of the most frequent complications associated with advanced malignancy is metastatic disease, particularly to bone.2 Though the exact incidence is unknown, approximately 210,00 to 400,000 new cases of metastatic bone disease are diagnosed annually, and roughly 350,000 people die annually with bone metastasis in the USA.2,3 Additionally, bone metastases can cause skeletal related events (SREs), such as pathological fractures, severe pain, compression of nearby neurovascular structures, and decreased mobility, which may negatively impact overall patient quality of life or require surgical intervention.4
Despite the significant impact of bone metastases on patient morbidity and mortality, their incidence by primary tumour and surgical outcomes remains poorly defined in contemporary oncological literature. Currently, it is estimated that up to 70% of all bone metastases are a result of metastatic breast and prostate cancer, with lung, kidney, and thyroid tumours following in frequency.5,6 The most common locations of bone metastasis are thought to be the spine, ribs, and pelvic bones.7 These estimates, however, may be skewed due to the paucity of comprehensive population-wide studies.8 Furthermore, modern advances in cancer treatment may not be reflected in these historic approximations.9 Recent investigations suggest the incidence of bone metastases from specific primary tumours may be evolving over time.10
In addition, postoperative survival and complication rates following surgical intervention for bone metastases remain difficult to assess. In a systematic review of studies examining surgical treatment of bone metastasis, Wood et al11 found a 17% complication rate and a 4% overall mortality rate for patients treated surgically for metastatic bone disease.11 These figures were limited by heterogeneity of primary tumour, differences in fixation method, and quality of evidence between studies.11 Prognostic factors have been proposed for postoperative survival, including preoperative anaemia and hypoalbuminemia and predictive scores such as the New England Spinal Metastasis Score, which attempt to characterize postoperative morbidity and mortality for specific sites of metastasis.12,13 While the clinical support tool PATHFx (Prognostics AB, Sweden) has made significant advancements in predicting postoperative survival in these patients, there is still work to be done in externally validating outcomes data across all patient populations.14
Therefore, the primary aim of this study is to identify the prevalence of various primary tumour subtypes in a population of patients with bone metastases. The secondary aim is to investigate whether rates of pathological fracture and mortality following surgical intervention for bone metastases vary depending on primary tumour subtype. Finally, this study describes differences in patient characteristics and postoperative outcomes for patients who underwent surgical intervention for an impending fracture versus those with a completed fracture.
Methods
A retrospective review was performed by querying all unique patients who were diagnosed with metastatic bone disease from 2015 to 2020 using the Premier Healthcare Database (PHD). PHD is a hospital administrative payer database in the USA that comprises inpatient and outpatient health information from diverse health systems across the country. The PHD contains granular, patient-level data regarding patient demographic details, hospital factors, insurance status, International Classification of Disease (ICD) diagnosis codes, ICD procedural codes, Current Procedural Terminology (CPT) codes, medication administration, and services rendered in the inpatient setting. The PHD is a nationally representative database that includes patients from over 1,000 hospitals and hospital systems across the USA. Patients with a diagnosis of metastatic disease of the bone were identified using ICD Tenth Revision (ICD-10)15 diagnosis code for secondary malignant neoplasm of the bone (C79.51). Only unique patients were included in this analysis, excluding repeat admissions under the same PHD-specific patient identifier.
This study was exempt from institutional review board review as all patient information was anonymized in accordance with the Health Insurance Portability and Accountability Act.
Identification of study cohorts
The overall cohort of patients was further categorized on the basis of their primary malignancy. All primary malignancies encoded by the ICD-10 diagnosis coding system were assessed, including oropharyngeal, gastrointestinal, lung, primary bone, skin, mesothelial, breast, prostatic, female reproductive, urinary, nervous system, endocrine, neuroendocrine, and haematogenous. These categories were further divided into malignancies of specific organs/organ systems based on clinical relevance.
Primary and secondary outcomes
The primary outcome investigated in this study was the prevalence of primary malignancies among the cohort of patients with secondary bone metastases. This was calculated by identifying the proportion of all patients with a secondary bone malignancy who had a concurrent ICD-10 diagnosis code for any of the aforementioned primary malignancies. These proportions were aggregated across all study years. The secondary outcomes included rates of pathological long bone fracture as well as 90-day and 360-day mortality rates following surgical treatment related to metastatic bone disease. Surgical treatment was defined as either treatment of an impending pathological long bone fracture or treatment of a completed pathological long bone fracture.
A subanalysis was performed to assess patient demographics, hospital factors, rates of medical comorbidities, 90-day risks, and 360-day risks of postoperative outcomes for patients who underwent surgery for impending pathological fractures compared to those who underwent surgery after completed pathological fractures. These two cohorts were identified depending on whether a ICD-10 diagnosis code for pathological fracture was present at the time of their admission.
Statistical analysis
All patient demographic details, hospital factors, comorbidities, rates of postoperative outcomes, and prevalence of primary tumours were presented using descriptive statistics. Chi-squared and independent-samples t-tests were used for categorical and continuous variables, respectively, to assess differences between patients treated for impending pathological fractures and those with completed fractures. Univariate regression analyses were performed to assess the 90-day risk of postoperative outcomes following surgery for the impending pathological fracture and completed fracture subanalysis. A multivariate model was designed to assess for potential confounders, including patient demographics, hospital factors, and medical comorbidities that approached a significant difference between cohorts (p < 0.100). Significance was defined as p < 0.05. All analyses were performed using STATA v. 16.1 (StataCorp).
Results
Primary cancer type
From 2015 to 2020, 407,893 unique patients with metastatic bone disease were identified and 14 primary tumour categories were assessed. The top five primary tumours were lung (24.8%), prostatic (19.4%), breast (19.3%), gastrointestinal (9.4%), and urological (6.5%) malignancies. (Table I) Uncategorized primary tumours comprised 5.7% of bone metastases (Table I).
Table I.
Prevalence of primary tumours with subtypes.
| Primary tumour | n (%) |
|---|---|
| Total | 407,893 |
| Oropharyngeal | 5,720 (1.4) |
| Gastrointestinal | 38,155 (9.4) |
| Upper GI | 8,736 (22.9) |
| Central GI | 6,308 (16.5) |
| Lower GI | 15,039 (39.4) |
| Hepatobiliary | 8,125 (21.3) |
| Lung | 101,279 (24.8) |
| Upper respiratory | 1,657 (1.6) |
| Lower respiratory | 98,846 (97.5) |
| Other | 776 (0.8) |
| Bone sarcoma | 3,083 (0.8) |
| Skin | 8,572 (2.1) |
| Melanoma | 5,372 (62.7) |
| Non-melanoma | 3,232 (37.7) |
| Mesothelial | 4,356 (1.1) |
| Breast | 78,658 (19.3) |
| Female reproductive | 9,059 (2.2) |
| Endometrial | 6,215 (68.6) |
| Other | 2,997 (33.1) |
| Male reproductive | 78,991 (19.4) |
| Prostate | 78,304 (99.1) |
| Non-prostate | 718 (0.9) |
| Urinary | 26,493 (6.5) |
| Renal | 16,379 (61.8) |
| Non-renal | 10,355 (39.1) |
| Central nervous system | 2,603 (0.6) |
| Endocrine | 3,843 (0.9) |
| Thyroid | 2,762 (71.9) |
| Non-thyroid | 1,083 (28.2) |
| Hematogenous | 19,481 (4.8) |
| Myeloma | 10,203 (52.4) |
| Lymphoma | 9,435 (48.4) |
| Neuroendocrine | 4,262 (1.0) |
| Uncategorized | 23,338 (5.7) |
Of the primary tumour subtypes, prostatic malignant tumours were associated with the oldest patients (mean 73.9 years (standard deviation (SD) 10.2) while central nervous system cancers were associated with the youngest patients (mean 57.5 years (SD 19.9)) (Table II). Excluding prostatic malignancies (37.6%), lung (26.0%), gastrointestinal (11.7%), renal (5.5%), and non-renal urinary (3.9%) tumours were the four most common malignant tumours occurring in male patients. Excluding breast tumours (44.3%), lung (26.7%), gastrointestinal (7.8%), female reproductive (5.2%), and renal (2.8%) tumours were the top four in female patients. The most common tumour subtypes by race were: lung in Asian patients (30.1%), prostatic in Black patients (26.7%), lung in Caucasian patients (27.3%), and breast in Hispanic patients (22.5%) (Table II).
Table II.
Demographic details by primary tumour.
| Variable | Mean age, yrs (SD) | Sex, n (%) | Race, n (%) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Female | Male | Asian | Black | Other | Unknown | Caucasian | Hispanic | ||
| Total | 175,029 | 209,758 | 8,393 | 48,652 | 23,933 | 6,493 | 296,791 | 21,860 | |
| Oropharyngeal | 63.6 (12.4) | 1,496 (0.9) | 4,222 (2.0) | 177 (2.1) | 612 (1.3) | 425 (1.8) | 116 (1.8) | 4,390 (1.5) | 305 (1.4) |
| Gastrointestinal | 65.2 (12.0) | 13,588 (7.8) | 24,561 (11.7) | 1,107 (13.2) | 5,030 (10.3) | 2,814 (11.8) | 707 (10.9) | 28,497 (9.6) | 2,764 (12.6) |
| Lung | 67.7 (10.6) | 46,652 (26.7) | 54,591 (26.0) | 2,527 (30.1) | 11,017 (22.6) | 5,368 (22.4) | 1,465 (22.6) | 80,902 (27.3) | 4,092 (18.7) |
| Breast | 63.9 (13.2) | 77,620 (44.3) | 1,020 (0.5) | 1,728 (20.6) | 10,209 (21.0) | 5,067 (21.2) | 1,340 (20.7) | 60,314 (20.3) | 4,917 (22.5) |
| Bone | 58.7 (20.3) | 1,347 (0.8) | 1,707 (0.8) | 81 (1.0) | 307 (0.6) | 245 (1.0) | 63 (1.0) | 2,387 (0.8) | 300 (1.4) |
| Skin | 67.8 (13.7) | 2,819 (1.6) | 5,752 (2.7) | 57 (0.7) | 210 (0.4) | 370 (1.5) | 80 (1.2) | 7,855 (2.6) | 328 (1.5) |
| Mesothelial | 60.2 (18.6) | 1,944 (1.1) | 2,410 (1.1) | 100 (1.2) | 588 (1.2) | 342 (1.4) | 83 (1.3) | 3,243 (1.1) | 360 (1.6) |
| Female reproductive | 62.4 (13.2) | 9,059 (5.2) | 0 (0.0) | 222 (2.6) | 1,314 (2.7) | 676 (2.8) | 188 (2.9) | 6,659 (2.2) | 752 (3.4) |
| Prostate | 74.1 (9.7) | 8 (0.01) | 78,925 (37.6) | 1,343 (16.0) | 12,986 (26.7) | 4,974 (20.8) | 1,484 (22.9) | 57,517 (19.4) | 4,318 (19.4) |
| Renal | 66.6 (11.6) | 4,928 (2.8) | 11,444 (5.5) | 285 (3.4) | 1,394 (2.9) | 1,050 (4.4) | 247 (3.8) | 13,403 (4.5) | 1,196 (5.5) |
| Non-renal urological | 71.6 (10.7) | 2,225 (1.3) | 8,129 (3.9) | 169 (2.0) | 852 (1.8) | 543 (2.3) | 146 (2.2) | 8,645 (2.9) | 511 (2.3) |
| Central nervous system | 57.5 (19.9) | 1,363 (0.8) | 1,240 (0.6) | 69 (0.8) | 239 (0.5) | 267 (1.1) | 55 (0.8) | 1,973 (0.7) | 214 (1.0) |
| Endocrine | 60 (20.4) | 2,025 (1.2) | 1,816 (0.9) | 143 (1.7) | 495 (1.0) | 326 (1.4) | 93 (1.4) | 2,786 (0.9) | 354 (1.6) |
| Myeloma | 68.6 (11.1) | 4,388 (2.5) | 5,814 (2.8) | 173 (2.1) | 1,979 (4.1) | 695 (2.9) | 201 (3.1) | 7,155 (2.4) | 659 (3.0) |
| Lymphoma | 68.7 (14.3) | 3,707 (2.1) | 5,727 (2.7) | 133 (1.6) | 919 (1.9) | 541 (2.3) | 163 (2.5) | 7,679 (2.6) | 570 (2.6) |
| Neuroendocrine | 65.6 (11.8) | 1,860 (1.1) | 2,400 (1.1) | 79 (0.9) | 501 (1.0) | 221 (0.9) | 59 (0.9) | 3,402 (1.1) | 220 (1.0) |
-
SD, standard deviation.
Primary tumour locations
The top five tumours with the highest rates of long bone pathological fracture were renal (5.8%), myeloma (3.4%), female reproductive (3.2%), lung (2.8%), and breast (2.7%) (Table III). Renal (4.2%), myeloma (3.0%), breast (2.4%), lung (2.4%), and bone (2.1%) malignancies required surgery for metastatic bone disease most frequently (Table III).
Table III.
Complications by primary tumour.
| Variable | Total, n | Pathological fracture, n (%) | Surgery, n (%) | 90-day mortality, n (%) | 1-year mortality, n (%) |
|---|---|---|---|---|---|
| Oropharyngeal | 5,720 | 103 (1.8) | 87 (1.5) | 5 (5.8) | 5 (5.8) |
| Gastrointestinal | 38,155 | 842 (2.2) | 607 (1.6) | 61 (10.1) | 70 (11.5) |
| Lung | 101,279 | 2,855 (2.8) | 2,396 (2.4) | 289 (12.1) | 333 (13.9) |
| Breast | 78,685 | 2,085 (2.7) | 1,883 (2.4) | 107 (5.7) | 132 (7.0) |
| Bone | 3,083 | 78 (2.5) | 64 (2.1) | 3 (4.7) | 3 (4.7) |
| Skin | 8,572 | 162 (1.9) | 121 (1.4) | 11 (9.1) | 15 (12.4) |
| Mesothelial | 4,356 | 101 (2.3) | 64 (1.5) | 1 (1.6) | 1 (1.6) |
| Female reproductive | 9,059 | 285 (3.2) | 124 (1.4) | 11 (8.9) | 15 (12.1) |
| Prostate | 78,304 | 1,249 (1.6) | 1,187 (1.5) | 59 (5.0) | 76 (6.4) |
| Renal | 16,379 | 951 (5.8) | 689 (4.2) | 47 (6.8) | 57 (8.3) |
| Non-renal urological | 10,355 | 269 (2.6) | 150 (1.5) | 15 (10.0) | 18 (12.0) |
| Central nervous system | 2,603 | 22 (0.9) | 19 (0.7) | 2 (10.5) | 2 (10.5) |
| Endocrine | 3,843 | 105 (2.7) | 80 (2.1) | 5 (6.3) | 8 (10.0) |
| Myeloma | 10,203 | 347 (3.4) | 308 (3.0) | 18 (5.8) | 23 (7.5) |
| Lymphoma | 9,435 | 222 (2.4) | 164 (1.7) | 17 (10.4) | 18 (11.0) |
| Neuroendocrine | 4,262 | 47 (1.1) | 44 (1.0) | 4 (9.1) | 5 (11.4) |
The 90-day mortality rates following treatment of pathological fracture of the long bones were greatest in lung (12.1%), central nervous system (10.5%), lymphoma (10.4%), gastrointestinal (10.1%), skin (9.1%), and neuroendocrine (9.1%) malignancies (Table III). Lung (13.9%), skin (12.4%), female reproductive system (12.1%), non-renal urinary (12.0%), and gastrointestinal (11.5%) tumours were the five primary tumours most likely to result in one-year mortality following treatment of pathological fracture of the long bones.
Timing of surgery
Of the 407,893 patients with bony metastases, 10,878 (3.7%) required surgery for an impending or completed pathological fracture. Fixation of impending pathological fractures was performed in 6,007 patients, while fixation for completed fractures was performed in 4,871. Patients in the impending fracture cohort were older (68.6 years (SD 12.1) vs 66.9 years (SD 11.6); p < 0.001, independent-samples t-test), more likely to be male (45.2% vs 42.5%; p = 0.004, chi-squared test), and had shorter postoperative length of stay (6.6 years (SD 7.4) vs 8.1 years (SD 8.1); p < 0.001, independent-samples t-test) than those in the completed group (Table IV). Impending fracture patients were more likely to be Caucasian (82.6% vs 78.5%; p < 0.001, chi-squared test) and more likely than those with completed fractures to be insured by Medicare (65.7% vs 60.8%; p < 0.001, chi-squared test). Patients with completed fractures were more likely than those with impending fractures to get treated at a teaching hospital and large hospitals with more than 500 beds (p < 0.001, chi-squared test).
Table IV.
Impending and completed fracture demographics.
| Variable | Impending | Completed | p-value |
|---|---|---|---|
| Total | 6,007 | 4,871 | |
| Mean age, yrs (SD) | 68.6 (12.1) | 66.9 (11.6) | < 0.001* |
| Mean LOS, days (SD) | 6.6 | 8.1 | < 0.001* |
| Total cost, USD (SD) | 24,860.26 (22,867.15) | 29,228.49 (25,267.16) | < 0.001* |
| Male, n (%) | 2,715 (45.2) | 2,069 (42.5) | 0.004† |
| Race, n (%) | |||
| Asian | 81 (1.4) | 86 (1.8) | < 0.001† |
| Black | 520 (8.7) | 512 (10.5) | |
| Other | 373 (6.2) | 370 (7.6) | |
| Unknown | 73 (1.2) | 79 (1.6) | |
| Caucasian | 4,960 (82.6) | 3,824 (78.5) | |
| Hispanic | 283 (4.7) | 279 (5.7) | 0.025† |
| Payer category, n (%) | |||
| Managed care organization | 998 (16.6) | 852 (17.5) | < 0.001† |
| Medicare | 3,949 (65.7) | 2,960 (60.8) | |
| Medicaid | 466 (7.8) | 508 (10.4) | |
| Other | 594 (9.9) | 551 (11.3) | |
| Marital status, n (%) | |||
| Married | 2,924 (48.7) | 2,306 (47.3) | 0.127† |
| Other | 577 (9.6) | 463 (9.5) | |
| Single | 2,490 (41.5) | 2,078 (42.7) | |
| Bed size, n (%) | |||
| < 100 | 209 (3.5) | 170 (3.5) | < 0.001† |
| 100 to 199 | 606 (10.1) | 398 (8.2) | |
| 200 to 299 | 880 (14.7) | 605 (12.4) | |
| 399 to 399 | 895 (14.9) | 685 (14.1) | |
| 400 to 499 | 792 (13.2) | 565 (11.6) | |
| > 500 | 2,625 (43.7) | 2,448 (50.3) | |
| Urban vs rural, n (%) | |||
| Rural | 601 (10) | 496 (10.2) | 0.760† |
| Urban | 5,406 (90) | 4,375 (89.8) | |
| Teaching status, n (%) | |||
| No | 2,550 (42.5) | 1,910 (39.2) | 0.001† |
| Yes | 3,457 (57.6) | 2,961 (60.8) | |
| Region, n (%) | < 0.001† | ||
| Midwest | 1,439 (24) | 984 (20.2) | |
| Northeast | 1,073 (17.9) | 1,048 (21.5) | |
| South | 2,606 (43.4) | 2,181 (44.8) | |
| West | 889 (14.8) | 658 (13.5) |
-
*
Independent-samples t-test.
-
†
Chi-squared test.
-
LOS, length of stay; SD, standard deviation.
Comorbidities based on surgery timing
Of the 27 comorbidities examined, statistically significant but quantitatively unremarkable differences in six were identified. Impending fracture patients had higher prevalence of congestive heart failure (10.6% vs 9.1%; p = 0.014, chi-squared test) than completed fracture fixation patients (Table V). Completed fracture fixation patients had higher prevalence of diabetes mellitus without complications (13.0% vs 11.8%; p = 0.046), drug abuse (3.8% vs 2.8%; p = 0.002), fluid imbalance (32.8% vs 28.4%; p < 0.001), obesity (15.9% vs 13.7%; p = 0.001), and weight loss (15.4% vs 13.1%; p = 0.001, all chi-squared test) than impending fracture patients (Table V).
Table V.
Impending and completed fracture comorbidities.
| Variable | Impending, n (%) | Completed, n (%) | p-value* |
|---|---|---|---|
| Total | 6,007 | 4,871 | |
| AKI/CKD | 885 (14.7) | 703 (14.4) | 0.659 |
| Alcohol abuse | 137 (2.3) | 120 (2.5) | 0.532 |
| Blood loss anaemia | 106 (1.8) | 78 (1.6) | 0.511 |
| CHF | 624 (10.6) | 445 (9.1) | 0.014 |
| Chronic PUD | 27 (0.5) | 27 (0.6) | 0.439 |
| Coagulopathy | 555 (9.2) | 462 (9.5) | 0.662 |
| COPD | 1,392 (23.2) | 1,080 (22.2) | 0.215 |
| Iron deficiency anaemia | 291 (4.8) | 274 (5.6) | 0.068 |
| Depression | 902 (15.0) | 723 (14.8) | 0.801 |
| DM without complications | 707 (11.8) | 635 (13) | 0.046 |
| DM with complications | 665 (11.1) | 587 (12.1) | 0.111 |
| Drug abuse | 166 (2.8) | 186 (3.8) | 0.002 |
| Fluid imbalance | 1,706 (28.4) | 1,596 (32.8) | < 0.001 |
| HIV/AIDS | 0 (0.2) | 11 (0.2) | 0.358 |
| Hypertension | 2,746 (45.7) | 2,236 (45.9) | 0.842 |
| Hypothyroidism | 949 (15.8) | 759 (15.6) | 0.758 |
| Liver disease | 269 (4.5) | 215 (4.4) | 0.872 |
| Lymphoma | 220 (3.7) | 181 (3.7) | 0.883 |
| Obesity | 824 (13.7) | 775 (15.9) | 0.001 |
| Other neurological disorder | 489 (8.1) | 365 (7.5) | 0.212 |
| Paralysis | 72 (1.2) | 78 (1.6) | 0.073 |
| Peripheral vascular disease | 374 (6.2) | 263 (5.4) | 0.068 |
| Psychosis | 38 (0.6) | 35 (0.7) | 0.585 |
| Pulmonary circulation disorder | 295 (4.9) | 261 (5.4) | 0.292 |
| Rheumatoid arthritis | 112 (1.9) | 71 (1.5) | 0.101 |
| Valve disease | 59 (1.0) | 34 (0.7) | 0.109 |
| Weight loss | 785 (13.1) | 750 (15.4) | 0.001 |
-
*
Chi-squared test.
-
AKI, acute kidney injury; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; PUD, peptic ulcer disease.
Complications based on surgery timing
After accounting for potentially confounding factors, patients who underwent completed fracture fixation had increased risk for pulmonary embolism (PE) (adjusted odds ratio (aOR) 1.31 (95% confidence interval (CI) 1.1 to 1.6); p = 0.010), deep vein thrombosis (DVT) (aOR 1.4 (95% CI 1.1 to 1.7); p = 0.003), urinary tract infection (UTI) (aOR 1.16 (95% CI 1.0 to 1.3); p = 0.004), transfusion (aOR 1.2 (95% CI 1.1 to 1.3) p = 0.001), 90-day mortality (aOR 1.2 (95% CI 1.0 to 1.3); p = 0.045), and 360-day mortality (aOR 1.1 (95% CI 1.0 to 1.3); p = 0.042, all multivariate analysis) compared to patients who underwent prophylactic surgery (Table VI).
Table VI.
Impending and completed fracture complications.
| Variable | Impending, n (%) | Completed, n (%) | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|---|---|
| OR (95% CI) | p-value | aOR (95% CI) | p-value | |||
| Total | 6,007 | 4,871 | ||||
| SSI | 19 (0.3) | 20 (0.4) | 1.3 (0.69 to 2.44) | 0.415 | 1.36 (0.72 to 2.57) | 0.340 |
| Sepsis | 407 (6.8) | 344 (7.1) | 1.05 (0.90 to 1.21) | 0.557 | 0.97 (0.83 to 1.13) | 0.700 |
| PE | 187 (3.1) | 206 (4.2) | 1.37 (1.12 to 1.68) | 0.002 | 1.31 (1.07 to 1.60) | 0.010 |
| DVT | 184 (3.1) | 211 (4.3) | 1.43 (1.17 to 1.75) | < 0.001 | 1.36 (1.11 to 1.67) | 0.003 |
| Wound dehiscence | 28 (0.5) | 25 (0.5) | 1.1 (0.64 to 1.89) | 0.726 | 0.96 (0.55 to 1.66) | 0.879 |
| Seroma | 7 (0.1) | 4 (0.1) | 0.7 (0.21 to 2.41) | 0.576 | 0.68 (0.20 to 2.35) | 0.542 |
| Stroke | 84 (1.4) | 67 (1.4) | 0.98 (0.71 to 1.36) | 0.919 | 0.97 (0.70 to 1.34) | 0.850 |
| Pneumonia | 324 (5.4) | 304 (6.2) | 1.17 (0.99 to 1.37) | 0.05 | 1.12 (0.95 to 1.32) | 0.182 |
| Respiratory failure | 553 (9.2) | 457 (9.4) | 1.02 (0.90 to 1.16) | 0.753 | 0.99 (0.86 to 1.13) | 0.887 |
| MI | 61 (1.0) | 50 (1.0) | 1.01 (0.69 to 1.47) | 0.955 | 1.15 (0.78 to 1.69) | 0.487 |
| AKI | 737 (12.3) | 650 (13.3) | 1.1 (0.98 to 1.23) | 0.095 | 1.09 (0.96 to 1.23) | 0.172 |
| UTI | 517 (8.6) | 478 (9.8) | 1.16 (1.01 to 1.32) | 0.03 | 1.16 (1.01 to 1.32) | 0.036 |
| Transfusion | 1,044 (17.4) | 1,033 (21.2) | 1.28 (1.16 to 1.41) | < 0.001 | 1.19 (1.08 to 1.32) | 0.001 |
| Acute blood loss anaemia | 1,708 (28.4) | 1,424 (29.2) | 1.04 (0.96 to 1.13) | 0.359 | 1.03 (0.95 to 1.13) | 0.443 |
| Haematoma | 23 (0.4) | 22 (0.5) | 1.18 (0.66 to 2.12) | 0.579 | 1.09 (0.60 to 1.97) | 0.776 |
| Haemorrhage | 13 (0.2) | 12 (0.3) | 1.14 (0.51 to 2.50) | 0.746 | 1.05 (0.47 to 2.33) | 0.912 |
| 90-day mortality | 397 (6.6) | 384 (7.9) | 1.21 (1.05 to 1.40) | 0.011 | 1.14 (1.01 to 1.33) | 0.045 |
| 360-day mortality | 485 (8.1) | 463 (9.5) | 1.2 (1.05 to 1.37) | 0.009 | 1.13 (1.02 to 1.30) | 0.042 |
| 90-day readmission | 690 (11.5) | 634 (13.0) | 1.15 (1.03 to 1.29) | 0.015 | 1.1 (0.98 to 1.24) | 0.098 |
-
AKI, acute kidney injury; aOR, adjusted odds ratio; CI, confidence interval; DVT, deep vein thrombosis; DVT, deep vein thrombosis; MI, myocardial infarction; PE, pulmonary embolism; SSI, surgical site infection; UTI, urinary tract infection.
Discussion
Bone is a common location for metastatic disease, and symptomatic bone metastases complicate the disease course of up to 40% of patients with the most common malignancies.6 In patients with breast and prostate cancer, bone metastases can be found in up to 70% of patients on post-mortem analysis.5 Skeletal metastases are associated with pathological fractures, hypercalcemia, spinal cord injury, and exacerbation of cancer-related pain, and can complicate the treatment course of the primary disease.16 Appropriate management of skeletal metastases requires multidisciplinary management, which includes an understanding of the primary tumour, disease course, potential for pathological fracture, and survival following surgical treatment.17 Studies prior to the widespread introduction of targeted therapy and immunotherapy consistently identified breast, kidney, lung, prostate, and thyroid malignancies to be the major contributors to metastatic disease of the skeleton.18-20 Our study demonstrates the primary tumours of contemporary patients with metastatic bone disease are most commonly lung, prostate, breast, gastrointestinal, and urinary malignancies. This shift, especially in the rise of gastrointestinal cancers metastatic to bone, is likely due, in part, to increased survival in this cohort thanks to immunotherapy and targeted therapy. As the number of patients with less common cancers metastatic to bone continues to rise, it is imperative that orthopaedic surgeons understand their survival profile and risk of pathological fracture, so as to best choose appropriate treatment options for these patients.21
Predicting survival following treatment of impending or completed pathological fracture has been historically challenging. Certain histologies, such as lung, have been associated with poorer survival, while others, such as breast, have longer life expectancy after treatment for skeletal metastases.22,23 However, these generalizations may not accurately predict disease trajectory for individual patients. There has been significant interest over the last decade in improving our predictive abilities, as this may help determine the most appropriate surgical treatment for each patient. Bayesian belief network models have been developed to determine survival after treatment of impending or completed pathological fracture and validated across multiple datasets in a number of countries.24-28 These networks, including the online clinical tool PATHFx, have since been applied to patients treated non-surgically for skeletal metastases as well.14 More recently, machine learning has been applied to this problem in a dataset of 1,090 patients with promising results.29 While not as individualized as these other methods, our cohort of over 400,000 patients and over 10,000 pathological fractures provides 90-day and 360-day mortality following pathological fracture treatment for 14 primary tumour categories. Of the 10,878 fractures analyzed, the impending fracture cohort included 6,007 patients and the completed fracture cohort included 4,871 patients, allowing for greater generalizability than any previous study. These data can be used to help prognosticate risk of pathological fracture and subsequent patient survival for each of these histological groups, and tailor appropriate surgical timing to the patient’s needs.
The advantage of prophylactic treatment of impending pathological fractures has been debated. Early studies demonstrated no difference in outcomes between impending and completed pathological fractures, both in terms of complications and survival.30 A modern cohort of 270 matched impending versus completed pathological fractures demonstrated no difference in 90-day survival or 30-day postoperative complications, despite higher intraoperative blood loss, anaesthesia time, and transfusions in the completed fracture cohort.31 Our study demonstrates improved 90- and 360-day survival in patients treated for impending pathological fracture compared to completed fracture, as well as significantly lower rates of DVT, PE, UTI, and blood transfusion, in a cohort ten times the size. Interestingly, 30-day postoperative complications after surgery for metastatic long bone disease has been associated with higher one-year mortality, making the avoidance of complications in this patient population even more important.32 In light of these findings, surgeons may feel more empowered to recommend prophylactic fixation for their patients with symptomatic metastatic bone disease, especially considering that this has also been demonstrated to be cost-effective.33
This study has multiple limitations that should be discussed. As with any large dataset, our data may be subject to coding errors, both in general and as a problem specifically identified in metastatic bone disease.34 Given the retrospective nature of these data, there may be unaccounted selection bias as to which patients received prophylactic fixation. This dataset does not have the ability to capture the patients’ outpatient therapy regimens, so no conclusions can be drawn as to whether certain complications could be related to common medical treatments for specific diseases. Additionally, disease burden to other organ systems could not be quantified. While this would affect survival, it is less likely to influence the treatment of pathological long bone fractures, except in the sickest cohort of patients at the very end of life. Finally, neoadjuvant or adjuvant radiation therapy could not be assessed, which is a mainstay of treatment for metastatic disease.
Despite these limitations, there are multiple strengths to this study. With a cohort size of over 400,000 patients and 10,878 pathological fractures, this is the largest study examining this clinical entity. Our large sample size allowed us to perform a robust analysis of all possible primary malignancies encoded by the ICD-10 coding system, and demonstrate a shift in the most common types of primary tumours seen in metastatic bone disease compared to historical series. Such data are important as clinicians assess disease type and trajectory, and help make decisions regarding metastatic disease care. This study was also able to retrospectively assess whether patients underwent prophylactic fixation of an impending fracture or surgery for a completed pathological fracture in a large patient population, which allowed for a robust assessment of postoperative outcomes. Finally, the PHD provides us the ability to examine mortality data of each tumour subtype on a granular scale. The use of this large-scale database to identify relative complication profiles and mortality among different tumour histologies will provide surgeons with an improved understanding of patient outcomes in those with impending fractures or completed fractures due to primary bone metastases.
This study defines the contemporary demographics of primary malignancies resulting in metastatic bone disease. It is important to understand the survival profile and risk for pathological fracture associated with each tumour subtype. Furthermore, patients with completed pathological fractures are more likely to experience postoperative complications, including 90-day and 360-day mortality. Physicians should consider these data when determining whether to choose prophylactic fixation of an impending fracture, and to prognosticate patient outcomes during treatment of their metastatic bone disease.
References
1. Ferlay J , Colombet M , Soerjomataram I . Cancer statistics for the year 2020: an overview . Int J Cancer . 2021 ; 149 ( 4 ): 778 – 789 . Crossref , PubMed Google Scholar
2. Weilbaecher KN , Guise TA , McCauley LK . Cancer to bone: a fatal attraction . Nat Rev Cancer . 2011 ; 11 ( 6 ): 411 – 425 . Crossref , PubMed Google Scholar
3. Huang J-F , Shen J , Li X , et al. Incidence of patients with bone metastases at diagnosis of solid tumors in adults: a large population-based study . Ann Transl Med . 2020 ; 8 ( 7 ): 482 . Crossref , PubMed Google Scholar
4. Coleman RE . Metastatic bone disease: clinical features, pathophysiology and treatment strategies . Cancer Treat Rev . 2001 ; 27 ( 3 ): 165 – 176 . Crossref , PubMed Google Scholar
5. Coleman RE . Clinical features of metastatic bone disease and risk of skeletal morbidity . Clin Cancer Res . 2006 ; 12 ( 20 Pt 2 ): 6243s – 6249s . Crossref , PubMed Google Scholar
6. Galasko C . The anatomy and pathways of skeletal metastases . In Weiss L , Gilbert AH . Bone Metastases . 1981 : 49 – 63 . Google Scholar
7. Kakhki VRD , Anvari K , Sadeghi R , Mahmoudian AS , Torabian-Kakhki M . Pattern and distribution of bone metastases in common malignant tumors . Nucl Med Rev Cent East Eur . 2013 ; 16 ( 2 ): 66 – 69 . Crossref , PubMed Google Scholar
8. Hernandez RK , Wade SW , Reich A , Pirolli M , Liede A , Lyman GH . Incidence of bone metastases in patients with solid tumors: analysis of oncology electronic medical records in the United States . BMC Cancer . 2018 ; 18 ( 1 ): 44 . Crossref , PubMed Google Scholar
9. Jensen AØ , Jacobsen JB , Nørgaard M , Yong M , Fryzek JP , Sørensen HT . Incidence of bone metastases and skeletal-related events in breast cancer patients: a population-based cohort study in Denmark . BMC Cancer . 2011 ; 11 : 29 . Crossref , PubMed Google Scholar
10. Ryan C , Stoltzfus KC , Horn S , et al. Epidemiology of bone metastases . Bone . 2022 ; 158 : 115783 . Crossref , PubMed Google Scholar
11. Wood TJ , Racano A , Yeung H , Farrokhyar F , Ghert M , Deheshi BM . Surgical management of bone metastases: quality of evidence and systematic review . Ann Surg Oncol . 2014 ; 21 ( 13 ): 4081 – 4089 . Crossref , PubMed Google Scholar
12. Schoenfeld AJ , Le HV , Marjoua Y , et al. Assessing the utility of a clinical prediction score regarding 30-day morbidity and mortality following metastatic spinal surgery: the New England Spinal Metastasis Score (NESMS) . Spine J . 2016 ; 16 ( 4 ): 482 – 490 . Crossref , PubMed Google Scholar
13. Kotian RN , Puvanesarajah V , Rao S , El Abiad JM , Morris CD , Levin AS . Predictors of survival after intramedullary nail fixation of completed or impending pathologic femur fractures from metastatic disease . Surg Oncol . 2018 ; 27 ( 3 ): 462 – 467 . Crossref , PubMed Google Scholar
14. Anderson AB , Wedin R , Fabbri N , Boland P , Healey J , Forsberg JA . External validation of PATHFx Version 3.0 in patients treated surgically and nonsurgically for symptomatic skeletal metastases . Clin Orthop Relat Res . 2020 ; 478 ( 4 ): 808 – 818 . Crossref , PubMed Google Scholar
15. World Health Organization . ICD-10: International Statistical Classification of Diseases and Related Health Problems: Tenth Revision . 2nd edition . Geneva, Switzerland : World Health Organization , 2004 . Google Scholar
16. D’Oronzo S , Coleman R , Brown J , Silvestris F . Metastatic bone disease: pathogenesis and therapeutic options: up-date on bone metastasis management . J Bone Oncol . 2019 ; 15 : 004 – 4 . Crossref , PubMed Google Scholar
17. Johnson CN , Gurich RW , Pavey GJ , Thompson MJ . Contemporary management of appendicular skeletal metastasis by primary tumor type . J Am Acad Orthop Surg . 2019 ; 27 ( 10 ): 345 – 355 . Crossref , PubMed Google Scholar
18. Gainor BJ , Buchert P . Fracture healing in metastatic bone disease . Clin Orthop Relat Res . 1983 ; 178 : 297 – 302 . PubMed Google Scholar
19. Habermann ET , Lopez RA . Metastatic disease of bone and treatment of pathological fractures . Orthop Clin North Am . 1989 ; 20 ( 3 ): 469 – 486 . PubMed Google Scholar
20. Mirels H . Metastatic disease in long bones. A proposed scoring system for diagnosing impending pathologic fractures . Clin Orthop Relat Res . 1989 ; 249):256-264 ( 249 ): 256 – 264 . PubMed Google Scholar
21. Jiang W , Rixiati Y , Zhao B , Li Y , Tang C , Liu J . Incidence, prevalence, and outcomes of systemic malignancy with bone metastases . J Orthop Surg (Hong Kong) . 2020 ; 28 ( 2 ): 2309499020915989 . Crossref , PubMed Google Scholar
22. Harrington KD . The management of acetabular insufficiency secondary to metastatic malignant disease . J Bone Joint Surg Am . 1981 ; 63-A ( 4 ): 653 – 664 . PubMed Google Scholar
23. Ghert M , Simunovic N , Cowan RW , Colterjohn N , Singh G . Properties of the stromal cell in giant cell tumor of bone . Clin Orthop Relat Res . 2007 ; 459 ( Suppl F ): 8 – 13 . Crossref , PubMed Google Scholar
24. Forsberg JA , Eberhardt J , Boland PJ , Wedin R , Healey JH . Estimating survival in patients with operable skeletal metastases: an application of a Bayesian belief network . PLoS One . 2011 ; 6 ( 5 ): e19956 . Crossref , PubMed Google Scholar
25. Forsberg JA , Wedin R , Bauer HCF , et al. External validation of the Bayesian Estimated Tools for Survival (BETS) models in patients with surgically treated skeletal metastases . BMC Cancer . 2012 ; 12 ( 1 ): 493 . Crossref , PubMed Google Scholar
26. Piccioli A , Spinelli MS , Forsberg JA , et al. How do we estimate survival? External validation of a tool for survival estimation in patients with metastatic bone disease-decision analysis and comparison of three international patient populations . BMC Cancer . 2015 ; 15 : 424 . Crossref , PubMed Google Scholar
27. Forsberg JA , Wedin R , Boland PJ , Healey JH . Can we estimate short- and intermediate-term survival in patients undergoing surgery for metastatic bone disease? Clin Orthop Relat Res . 2017 ; 475 ( 4 ): 1252 – 1261 . Crossref , PubMed Google Scholar
28. Ogura K , Gokita T , Shinoda Y , et al. Can a multivariate model for survival estimation in skeletal metastases (PATHFx) be externally validated using Japanese patients? Clin Orthop Relat Res . 2017 ; 475 ( 9 ): 2263 – 2270 . Crossref , PubMed Google Scholar
29. Thio Q , Karhade AV , Bindels BJ , et al. Development and internal validation of machine learning algorithms for preoperative survival prediction of extremity metastatic disease . Clin Orthop Relat Res . 2020 ; 478 ( 2 ): 322 – 333 . Crossref , PubMed Google Scholar
30. Wedin R , Bauer HC , Wersäll P . Failures after operation for skeletal metastatic lesions of long bones . Clin Orthop Relat Res . 1999 ; 358 ( 358 ): 128 – 139 . Crossref , PubMed Google Scholar
31. Groot OQ , Lans A , Twining PK , et al. Clinical outcome differences in the treatment of impending versus completed pathological long-bone fractures . J Bone Joint Surg Am . 2022 ; 104-A ( 4 ): 307 – 315 . Crossref , PubMed Google Scholar
32. Bindels BJJ , Thio Q , Raskin KA , Ferrone ML , Lozano Calderón SA , Schwab JH . Thirty-day postoperative complications after surgery for metastatic long bone disease are associated with higher mortality at 1 year . Clin Orthop Relat Res . 2020 ; 478 ( 2 ): 306 – 318 . Crossref , PubMed Google Scholar
33. Blank AT , Lerman DM , Patel NM , Rapp TB . Erratum to: Is prophylactic intervention more cost-effective than the treatment of pathologic fractures in metastatic bone disease? Clin Orthop Relat Res . 2016 ; 474 ( 7 ): 1742 – 1743 . Crossref , PubMed Google Scholar
34. Hanna SM , Ramsey DC , Doung YC , et al. Utility of the current procedural terminology codes for prophylactic stabilization for defining metastatic femur disease . J Am Acad Orthop Surg Glob Res Rev . 2020 ; 4 ( 12 ): e20 . Crossref , PubMed Google Scholar
Author contributions
A. B. Christ: Conceptualization, Methodology, Supervision, Investigation, Validation, Writing – original draft, Writing – review & editing.
A. S. Piple: Data curation, Formal analysis, Writing – original draft, Writing – review & editing.
B. S. Gettleman: Data curation, Formal analysis, Writing – original draft, Writing – review & editing.
A. Duong: Data curation, Formal analysis, Writing – original draft, Writing – review & editing.
M. Chen: Data curation, Formal analysis, Writing – original draft, Writing – review & editing.
J. C. Wang: Data curation, Formal analysis, Writing – original draft, Writing – review & editing.
N. D. Heckmann: Conceptualization, Supervision, Investigation, Formal analysis, Writing – original draft.
L. Menendez: Conceptualization, Supervision, Investigation, Formal analysis, Writing – original draft, Writing – review & editing.
Funding statement
The authors received no financial or material support for the research, authorship, and/or publication of this article.
ICMJE COI statement
A. B. Christ reports consulting fees from Intellijoint Surgical and Smith & Nephew, unrelated to this study. N. D. Heckmann reports royalties or licenses from Corin, consulting fees from Microport Orthopedics, Intellijoint Surgical, Corin, and Zimmer Biomet, and stock or stock options in Intellijoint Surgical, all of which are unrelated to this study. L Menendez reports Royalties or licenses from TeDen Surgical and consulting fees from Onkos Surigcal, all of which are unrelated to this study.
Data sharing
The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.
Ethical review statement
The data used for this study came from a de-identified payer database; therefore, IRB review was not required.
Open access funding
The authors confirm that the open access fee was self-funded.
Follow A. B. Christ @AlexChristMD
© 2023 Authors et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/
