- Email: [email protected]
Fracture after gastrectomy for gastric cancer: A long-term follow-up observational study
European Journal of Cancer 72 (2017) 28e36
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Original Research
Fracture after gastrectomy for gastric cancer: A long-term follow-up observational study Hyun Jin Oh a, Chul-Hyun Lim a,*, Byung-Ho Yoon b, Seung Bae Yoon a, Myong Ki Baeg a, Won Chul Kim a, Yu Kyung Cho a, Jae Myung Park a, Myung-Gyu Choi a, Han Mo Yoo c, Kyo Young Song c, Hae Myung Jeon c, Cho Hyun Park c a Division of Gastroenterology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Republic of Korea b Department of Orthopaedic Surgery, Inje University College of Medicine, Seoul Paik Hospital, Republic of Korea c Division of Gastrointestinal Surgery, Department of Surgery, College of Medicine, The Catholic University of Korea, Republic of Korea
Received 26 May 2016; received in revised form 8 November 2016; accepted 16 November 2016
KEYWORDS Gastric cancer; Gastrectomy; Osteoporosis; Fracture; Risk factor
Abstract Aim: Gastrectomy is a known risk factor for decreased bone mass. We aimed to evaluate the cumulative incidence and predictive factors of fracture in gastric cancer patients who underwent gastrectomy. Method: We retrospectively reviewed the records of 1687 patients who underwent gastrectomy for gastric cancer at our hospital between September 1991 and December 2008. The exclusion criteria were stage IV gastric cancer, history of cancer recurrence, medical conditions that cause osteoporosis and high-energy injury. Fractures at sites considered to be associated with osteoporosis were diagnosed radiologically. Results: In total, our analysis included the records of 1131 patients. The incidence of postgastrectomy fracture was 42.1 cases per 1000 person-years. Fractures typically occurred within 3.7 0.5 years postoperatively. The cumulative incidence of fracture was 9.1%, 19.7%, and 37.3% by postoperative year 2, 4, and 6, respectively. During the following years, the cumulative incidence increased slowly, up to a final 40.6%. Multivariate analysis showed that older age (hazard ratio, 1.03; 95% confidence interval, 1.01e1.04) and smoking (hazard ratio, 1.35; 95% confidence interval, 1.05e1.73) were significantly associated with fracture, whereas sex, body mass index, percent weight loss, diabetes mellitus, tumour stage, and type of gastrectomy were not.
* Coresponding author: Division of Gastroenterology, Department of Internal Medicine, College of Medicine, Seoul St. Mary’s Hospital, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea. Fax: þ82 2 2258 2089. E-mail address: [email protected] (C.-H. Lim). http://dx.doi.org/10.1016/j.ejca.2016.11.023 0959-8049/ª 2016 Elsevier Ltd. All rights reserved.
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
29
Conclusion: The cumulative incidence of fracture is high in gastric cancer patients who have undergone gastrectomy, and fracture rate is higher during the early postoperative period. Old age and smoking are independent risk factors for postgastrectomy fracture in these patients. More detailed postoperative surveillance and pharmacological intervention should be considered to prevent fracture. ª 2016 Elsevier Ltd. All rights reserved.
1. Introduction Gastric cancer is the 4th most common type of cancer worldwide [1], and the incidence of the disease is especially high in East Asia. Early diagnosis and curative treatment such as gastrectomy can improve the outcomes. In Korea, the survival rate for gastric cancer is over 90% when diagnosed early [2]. However, gastrectomy is a known risk factor for osteoporosis [3e5], and although the pathogenesis has not been fully elucidated, the reported incidence of osteoporosis after gastrectomy ranges from 32% to 42% [3,6,7], which implies a higher risk of fractures. Fractures decrease quality of life, and increase the economic burden on individuals, as well as on society [8e10]. In a large cohort study involving peptic ulcer patients, the incidence of osteoporotic fracture after gastrectomy was up to 70% [6]. However, to date, the incidence, characteristics, and risk factors associated with fracture after gastrectomy have not been studied longitudinally and systemically in a large cohort of gastric cancer patients. Thus, the present study aimed to evaluate the cumulative incidence and predictive factors of postgastrectomy fracture in gastric cancer patients who underwent gastrectomy.
relevant data, including clinical, surgical, and pathological records, were collected retrospectively from the database. The Institutional Review Board of our institution approved this study (KC16RISI0380). 2.2. Follow-up measurements All included patients were registered and followed-up by the Comprehensive Cancer Institute of Seoul St. Mary’s Hospital. The patients were followed-up every 3 months during the first two postoperative years, every 6 months during the next three postoperative years, and annually after the fifth postoperative year. The follow-up program consisted of interim history taking, physical examination, imaging investigations, endoscopic examination, haematology tests and blood chemistry panels. Imaging investigations included chest radiography, abdominal computerised tomography (CT) and bone scintigraphy. Clinical and laboratory parameters regarding nutritional status, including body mass index (BMI), serum albumin, and serum calcium levels were reviewed at baseline and at one year postoperatively. The percentage of weight loss was calculated as the difference between the baseline and 1-year body weight, divided by the baseline body weight.
2. Patients and methods 2.3. Outcomes measured 2.1. Patients From September 1991 to December 2008, a total of 1687 patients with gastric cancer underwent curative gastrectomy at the Seoul St. Mary’s Hospital, Seoul, Korea. Gastrectomy was defined as margin-negative surgical resection, with no residual tumour (macroscopically). The exclusion criteria were: age >70 years at gastrectomy regarded as decreasing calcium absorption (n Z 334) [11,12], postoperative tumour recurrence (n Z 118), another type of cancer (n Z 33). Among the remaining 1202 patients, those with another medical condition (n Z 26) or who underwent surgery (n Z 45) known to affect bone mineral density were also excluded. Therefore, the records of a total of 1131 patients were included in the final analysis (Fig. 1). Cancers were staged in accordance with the tumour node metastasis classification system used by the Union for International Cancer Control, 6th edition [13]. All
Fracture was diagnosed by annual bone scintigraphy, with additional CT, magnetic resonance imaging, or Xray investigations performed as clinically indicated, at the site of bone pain with or without trauma, or as needed for the differential diagnosis from bone metastasis. All radiographs, bone scans and magnetic resonance imaging scans were analysed for fractures by trained radiologists. Only those fractures at sites considered to be associated with osteoporosis (i.e. spine, coccyx, pelvis, ribs, humerus, forearm, elbow, hip, femur, tibia, fibula, clavicle, scapula, and sternum) were considered in this study. The trauma history of the patients with fractures was investigated to differentiate low-energy trauma fracture (following fall from standing height or less) from high-energy trauma fracture (following fall from a height greater than standing height). Person-years were counted from enrolment to the date of death or the end of follow-up.
30
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 1. Flow chart of enrolment and outcomes.
2.4. Statistical analysis Pearson’s chi-square test and the Student t test for unpaired data were applied for categorical and continuous variables, respectively, to compare the groups of patients with and without postgastrectomy fracture with respect to clinicopathological characteristics. A P-value of <0.05 was regarded as significant. The cumulative incidence of postgastrectomy fracture was calculated by the KaplaneMeier method, with the date of gastrectomy as the starting point. Differences in incidence were examined by the log-rank test. Multivariate analysis was performed using a Cox proportional hazards model with a backward stepwise selection procedure. All analyses were performed using SAS for Windows (version 8.02, SAS Institute, Cary, NC, USA). 3. Results 3.1. Characteristics of the study population Our study involved 1131 patients, with 344 women (30.4%), of whom 193 were postmenopausal. The mean
age was 53.8 years (range, 20e69 years). Distal subtotal gastrectomy was the most commonly performed procedure (74.0%, n Z 837). The majority of tumours (74.9%, n Z 848) were evaluated as stage I (Table 1). The patients were followed-up for 9216 person-years. The mean follow-up period was 8.4 4.2 years. Of 1131 patients, 336 (29.7%) had postgastrectomy fracture. 3.2. Incidence of postgastrectomy fracture The incidence of postgastrectomy fracture was 42.1 cases per 1000 person-years. Fractures occurred on average 3.7 0.5 years postoperatively. KaplaneMeier analysis showed a cumulative incidence of 9.1%, 19.7%, and 37.3% by postoperative year 2, 4 and 6, respectively. During the following years, the incidence of fractures continued to increase slowly, up to a final 40.6% (Fig. 2A). The incidence of postgastrectomy fracture was 48.7 and 30.1 cases per 1000 person-years for males and females, respectively, with significantly higher incidence in males (P < 0.001; Fig. 2B). The incidence of postgastrectomy fracture was also higher in
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
31
Table 1 Baseline characteristics of patients. Measures Age (Mean SD) Male Smoking Diabetes mellitus Postmenopausal state Preoperative BMI (kg/m2, Mean SD) Anaemiaa Alkaline phosphatase (U/L) Type of surgery Total gastrectomy Subtotal gastrectomy Postoperative CTx Follow-up duration (month) Mean SD Range Tumour size (mm, Mean SD) T stage T1 T2 T3 T4 N stage N0 N1 N2 N3 Pathological stage I II III Histology Differentiated Undifferentiated NA
Total population (N Z 1131)
Postgastrectomy fracture Yes (N Z 336)
No (N Z 795)
53.8 9.9 787 (69.6%) 313 (27.7%) 68 (6.0%) 193 (193/344, 56.1%) 23.3 3.1 325 (28.7%) 118.07 64.3
55.8 9.1 255 (75.9%) 115 (34.4%) 25 (7.5%) 58 (58/81, 71.6%) 23.3 3.1 91 (27.1%) 121.0 73.3
52.9 10.2 532 (66.9%) 198 (25%) 43 (5.4%) 135 (135/263, 51.3%) 23.3 3.1 234 (29,5%) 117.7 60.1
294 (26.0%) 837 (74.0%) 218 (19.3%)
83 (24.7%) 253 (75.3%) 69 (20.8%)
211 (26.5%) 584 (73.5%) 149 (18.6%)
101.0 50.2 4e279 35.1 22.8
100.8 45.3 4e279 34.4 22.3
101.1 52.1 8e261 35.4 23.0
696 (61.5%) 328 (29.0%) 104 (9.2%) 3 (0.3%)
209 (62.0%) 94 (28.0%) 31 (9.3%) 2 (0.6%)
487 (61.2%) 234 (29.5%) 73 (9.3%) 1 (0.1%)
810 (71.6%) 151 (13.4%) 102 (9.0%) 68 (6.0%)
227 (67.6%) 50 (14.9%) 32 (9.5%) 27 (8.0%)
583 (73.3%) 101 (12.7%) 70 (8.8%) 41 (5.2%)
848 (74.9%) 194 (17.2%) 89 (7.9%)
246 (73.2%) 54 (16.1%) 36 (10.7%)
602 (75.7%) 140 (17.6%) 53 (6.7%)
527 (46.7%) 601 (53.1%) 3 (0.3%)
169 (50.3%) 167 (49.7%) 0 (0%)
358 (45.0%) 434 (54.6%) 3 (0.4%)
P-value <0.001 0.003 0.001 0.189 0.001 0.904 0.418 0.426 0.519
0.581 0.929
0.499 0.189
0.202
0.067
0.226
Where appropriate, data are shown as the mean SD. Abbreviations: BMI, body mass index; CTx, chemotherapy; NA, not applicable; SD, standard deviation. a Anaemia was defined by World Health Organisation criteria: Hb < 12 g/dL in women and <13 g/dL in men.
postmenopausal women than in premenopausal women (P Z 0.002; Fig. 2C). To assess the impact of BMI on fracture incidence, each patient was assigned to one of three BMI-based groups: underweight (<18.5 kg/m2), normal 2 (18.5e24.9 kg/m ) or overweight (25.0 kg/m2) [14]. This assignment was performed once for preoperative BMI, and again for postoperative BMI. The incidence of postgastrectomy fracture was similar among the preoperative BMI-based groups, as well as among the 1year postoperative BMI-based groups. Similarly, no significant association with the incidence of postgastrectomy fracture was found regarding percent weight loss during postoperative year 1 (weight loss group, 10%; stable weight group, <10%), or type of surgery (total or subtotal gastrectomy). 3.3. Location of postgastrectomy fractures Out of the 336 patients with postgastrectomy fracture (Table 2), 90.7% had solitary fracture, whereas 9.9% had
multiple fractures. The following types of solitary fractures occurred: rib fractures, in 180 patients (59.4% of all patients with solitary fracture), within 3.8 0.2 years postoperatively; spinal fractures, in 80 patients (26.4%), within 4.2 0.4 years postoperatively; hip fracture, 12 patients (4.1%), within 2.3 0.4 years postoperatively; wrist fracture, in 10 patients (3.3%), within 3.8 0.4 years postoperatively; and other fractures (humerus, clavicle, sternum and tibia), in 21 patients (6.9%). Of the 33 patients with multiple fractures, 17 had hip fractures, whereas 11 had vertebral fracture accompanied by fracture at other sites. Multiple fractures occurred within 3.7 0.5 years postoperatively. 3.4. Risk factors for postgastrectomy fracture Compared with the group of patients without fracture, fracture group contained older patients (55.8 9.1 years versus 52.9 10.2 years; P < 0.001), more males (75.9% versus 66.9%; P Z 0.003), more smokers (34.4% versus 25%; P < 0.001), and more postmenopausal women
32
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 2. Cumulative incidence of the postgastrectomy fracture in gastric cancer patients for (A) overall, (B) male and female and (C) postmenopausal and premenopausal females.
(71.6% versus 51.3%; P < 0.001; Table 1). There were no significant differences between the groups with respect to incidence of diabetes mellitus, preoperative BMI, type of surgery, postoperative chemotherapy treatment and tumour stage. Multivariate Cox analysis showed that older age was a significant factor associated with postgastrectomy fracture (hazard ratio, 1.03; 95% confidence interval, 1.01e1.04; Table 3). Among the other variables, smoking (hazard ratio, 1.35; 95% confidence interval, 1.05e1.73) was also an independent risk factor for postgastrectomy fracture, whereas sex, BMI, percent
weight loss, diabetes mellitus, tumour stage or type of gastrectomy were not significantly associated with the incidence of fracture. 3.5. Postgastrectomy changes related to nutritional status BMI and laboratory parameters indicating nutritional status were compared between the two groups of patients (Table 4). All recorded levels of calcium and albumin were within normal limits. At baseline, albumin levels were lower in patients with postgastrectomy fracture and had increased significantly in both groups
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Table 2
Table 4
Sites of postgastrectomy fracture. Anatomic site
Number of patients
Vertebra Ribs Femur Wrist Othersa
303 (90.1%) 80 180 12 10 21 33 (9.9%)
Solitary bone fracture
Multiple bone fracture a
33
Parameters of patients with postgastrectomy fracture and without postgastrectomy fracture. With fracture
Without fracture
P
4.0 0.4 4.3 0.5a
4.1 0.3 4.3 0.3a
<0.001 0.701
8.6 0.7 8.9 0.5a
8.9 0.5 8.9 0.4
<0.001 0.384
23.3 3.1 21.0 2.8a
23.3 3.1 21.0 2.8a
0.344 0.652
b
Including humerus, clavicle, sternum and tibia.
by postoperative year 1. Similarly, at baseline, calcium levels were significantly lower in patients with fracture, but, by postoperative year 1, had increased only in these patients. A significant decrease in BMI was found at 1 year postoperatively for both groups. 3.6. Postgastrectomy fracture and overall survival Fig. 3 shows the postoperative overall survival curves of patients according to the postgastrectomy fracture. The 5-year overall survival rate was not different between two groups (98.1% versus 97.8%). During the follow-up years, overall survival rate was lower in patients with postgastrectomy fracture (76.4% versus 73.4%). However, no significant difference was observed between two groups (P Z 0.880). 4. Discussion The present study aimed to evaluate the postgastrectomy fracture risk in gastric cancer patients. Our retrospective analysis reviewed 20 years of experience at a single tertiary institution, and demonstrated high cumulative incidence of postgastrectomy fracture in such patients, as we found radiologic and historical evidence of postgastrectomy fracture in 42.1 cases per 1000 person-years. Moreover, since 18.7% of the patients were followed-up for less than 5 years postoperatively, the incidence of fractures may be underestimated. Age and smoking were independent prognostic factors after adjustment for sex, menopausal state and cancer stage,
Albumin (g/dL) Baseline 1 year after surgery Calcium (mg/dL)b Baseline 1 year after surgery BMI Baseline 1 year after surgery
Where appropriate, data are shown as the mean SD. a P < .05 (paired t test; baseline versus 1 year after surgery). b Albumin and calcium were evaluated in 465 and 460 patients, respectively.
whereas baseline and 1-year postoperative BMIs were not associated with the incidence of postgastrectomy fracture. To the best of our knowledge, this is the first long-term follow-up observational cohort study to investigate the incidence and risk factors of postgastrectomy fracture in gastric cancer patients. Between 2005 and 2008, the incidence of osteoporosis-related fractures in the general Korean population aged >50 years was 1635 cases per 100,000 person-years [15], which is over 2.5 times lower than the incidence of postgastrectomy fracture we found in the cohort of gastric cancer patients. In males, we found a much higher incidence than that reported for the general population (48.7 cases per 1000 person-years versus 725 cases per 100,000 person-years); this trend was also true for females, but to a smaller extent (30.1 cases per 1000 person-years in our cohort, versus 2408 cases per 100,000 person-years in the general population). In contrast to the other studies [15e17], we found a higher incidence of fracture in males. This is likely related to the fact that gastric cancer is more frequent in males, and our cohort was predominantly male (70%). Moreover, the proportion of smokers was higher among males than among females (38.6% versus 2.6%), and we found that smoking was independently associated with the risk for postgastrectomy fracture. Smoking is a well
Table 3 Cox proportional hazard models for cumulative incidence of postgastrectomy fracture in gastric cancer patients. Variables
Age Male versus female Postmenopausal versus premenopausal BMI Body weight change (%) Diabetes mellitus Smoking Stage III versus stage I Subtotal gastrectomy versus. total gastrectomy
Univariate analysis
Multivariate analysis
HR
95% CI
P value
HR
95% CI
P value
1.03 1.58 2.16 1.01 1.00 1.30 1.43 1.51 1.08
1.02e1.04 1.23e2.03 1.33e3.50 0.97e1.04 0.99e1.01 0.87e1.96 1.14e1.79 1.06e2.14 0.84e1.39
<0.001 <0.001 0.002 0.786 0.929 0.202 0.002 0.021 0.528
1.03 1.15 1.54
1.01e1.04 0.35e3.76 0.92e2.57
<0.001 0.820 0.095
1.35 1.41
1.05e1.73 0.99e2.00
0.018 0.055
Abbreviations: HR, hazard ratio; CI, confidence interval; BMI, body mass index.
34
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 3. Overall survival according to the postgastrectomy fracture.
known risk factor for fracture in general population [16]. The Fracture Risk Assessment Tool is used in clinical settings to predict the probability of major fracture based on clinical risk factors such as BMI [18]. Body weight loss of 5e15% typically occurs after gastrectomy, and is maintained without significant changes until 1 year postoperatively [19,20]. In our study, the percent body weight loss was within these limits. In general, body weight loss or lower BMI have been known to increase the risk of fracture, but this association has recently been challenged. Our study suggested that percentage of body weight loss, baseline BMI, and 1year postoperative BMI are not associated with the incidence of fractures. Our results are consistent with the findings presented in recent reports for general population as well as patients with bariatric surgery [16,21e23]. In the patients included in the present study, the most common fracture sites were the ribs and the spine, which is similar to reports of fractures after liver transplantation [17]. In addition, our study included the incident osteoporotic fractures. In general, these asymptomatic fractures are relevant because of increasing risk of subsequent osteoporotic fractures [24e26]. In other large-population studies, such as the National Osteoporosis Risk Assessment involving postmenopausal women, or the Osteoporotic Fractures in Men study, rib fracture was the most common and important incident clinical fracture, and it was found that a history of rib fracture carried at least a 2-fold increased risk of an incident subsequent osteoporotic fracture [27,28]. The pathogenesis of bone loss and fracture in gastric cancer patients who underwent gastrectomy is uncertain. One possible explanation is malabsorption. Specifically, calcium is absorbed primarily in the duodenum, and gastrectomy with anastomosis alters normal gastrointestinal physiology. Gastric dumping and the formation of insoluble calcium soaps due to
malabsorption of fat may result in malabsorption of calcium. Decreased acid secretion in the gastric mucosa also may result in malabsorption of calcium. However, serum calcium levels of postgastrectomy patients were reported to be within normal limits [4,29,30], consistent with our data. Moreover, 25-hydroxyvitamin D levels are within normal or, at most, subnormal levels [30,31]. Calcium homoeostasis and elevated parathyroid (PTH) hormones after gastrectomy may explain the normal calcium levels and bone loss after gastrectomy. The mean serum PTH levels increase after gastrectomy [32,33], and patients with higher postoperative PTH level exhibited a more pronounced decrease in bone mass density (BMD). In our study, patients with postgastrectomy fracture exhibited a significant increase in serum calcium levels by 1 year postoperatively, which may reflect maintenance of serum calcium homoeostasis at the expense of BMD, subsequently leading to higher risk of fractures. Thus, a remodelling imbalance with dominant bone resorption appears to contribute to bone loss and fracture in these patients. There are no specific screening guidelines and no established risk factor grading system for bone health in gastric cancer patients. Although, based on insufficient data, American Gastroenterological Association recommended BMD measurement in gastrectomy patients [34], best surveillance schedule and method were not defined. For liver transplantation patients, during the first 5 postoperative years, BMD should be measured annually in osteopenic patients, and every 2e3 years in patients with normal BMD [35]. We found that fracture mostly occurred within the first 6 postoperative years, but the cumulative incidence continued to increase afterwards. In a prospective study involving gastrectomy patients, bone remodelling imbalance occurred early within the first postoperative year [33]. Therefore, bone pain or fracturing, serum calcium, phosphorous, 25hydroxyvitamin D and PTH levels should be assessed at least within 6 years postoperatively. Regular follow-
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
up tests and appropriate intervention should also be considered. Dominant bone resorption due to calcium homoeostasis and secondary hyperparathyroidism may also occur in the early postgastrectomy period, in which case dietary calcium and bisphosphonate can be administered prophylactically. In rats, bisphosphonate was effective in blocking gastrectomy-induced decreases in BMD of the lumbar spine and femoral bone, and in femoral bending strength [36]. Although comparable human studies are scarce, we believe bisphosphonate may be used to prevent bone resorption and fracture in gastric cancer patients who underwent gastrectomy, along with dietary calcium and vitamin D supplements even when serum calcium levels are within normal limits. The results of the present study are highly relevant because: the incidence and location of fractures was assessed based on radiologic reports and low trauma history; the analysis was longitudinal; the study population was large; we also investigated parameters particularly important in the target population, namely baseline and postoperative BMIs, as well as nutritionrelated laboratory parameters. This study is limited due to its retrospective nature, and the fact that we could not obtain serum markers for bone metabolism, 25hydroxyvitamin D, PTH, or BMD data. Although our study focused on the fracture risk of gastric cancer patients who underwent curative gastrectomy, it can be extended to the advanced gastric cancer in future study. In conclusion, the cumulative incidence of fracture is high in patients with gastric cancer undergone gastrectomy, and high rate of fracturing is reported at the early period of 6 years after gastrectomy. Old age and smoking are the independent risk factors for postgastrectomy fracture. More intensive surveillance pharmacologic interventions should be considered to prevent fracture. Funding and support This research did not receive any specific grant from funding agencies in the public, commercial or not-forprofit sectors. Conflict of interest statement None declared.
References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69e90. [2] Lee H-J, Yang H-K, Ahn Y-O. Gastric cancer in Korea. Gastric Cancer 2002;5:0177e82.
35
[3] Inoue K, Shiomi K, Higashide S, Kan N, Nio Y, Tobe T, et al. Metabolic bone disease following gastrectomy: assessment by dual energy X-ray absorptiometry. Br J Surg 1992;79:321e4. [4] Liedman B, Bosaeus I, Mellstrom D, Lundell L. Osteoporosis after total gastrectomy. Results of a prospective, clinical study. Scand J Gastroenterol 1997;32:1090e5. [5] Tovey FI, Hall ML, Ell PJ, Hobsley M. A review of postgastrectomy bone disease. J Gastroenterol Hepatol 1992;7: 639e45. [6] Pryor JP, O’Shea MJ, Brooks PL, Datar GK. The long-term metabolic consequences of partial gastrectomy. Am J Med 1971; 51:5e10. [7] Lim JS, Kim SB, Bang HY, Cheon GJ, Lee JI. High prevalence of osteoporosis in patients with gastric adenocarcinoma following gastrectomy. World J Gastroenterol 2007;13:6492e7. [8] Johnell O. The socioeconomic burden of fractures: today and in the 21st century. Am J Med 1997;103. 20Se5S; discussion 5Se6S. [9] Burge R, Dawson-Hughes B, Solomon DH, Wong JB, King A, Tosteson A. Incidence and economic burden of osteoporosisrelated fractures in the United States, 2005e2025. J Bone Min Res 2007;22:465e75. [10] Si L, Winzenberg TM, de Graaff B, Palmer AJ. A systematic review and meta-analysis of utility-based quality of life for osteoporosis-related conditions. Osteoporos Int 2014;25:1987e97. [11] Bullamore J, Wilkinson R, Gallagher J, Nordin B, Marshall D. Effect of age on calcium absorption. Lancet 1970;296:535e7. [12] Riggs BL. Overview of osteoporosis. West J Med 1991;154:63. [13] Sobin LH, Fleming ID. TNM classification of malignant tumors. Cancer 1997;80:1803e4. [14] WHO EC. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157. [15] Park C, Ha YC, Jang S, Jang S, Yoon HK, Lee YK. The incidence and residual lifetime risk of osteoporosis-related fractures in Korea. J Bone Min Metab 2011;29:744e51. [16] Crandall CJ, Yildiz VO, Wactawski-Wende J, Johnson KC, Chen Z, Going SB, et al. Postmenopausal weight change and incidence of fracture: post hoc findings from women’s health initiative observational study and clinical trials. BMJ 2015;350: h25. [17] Guichelaar MM, Schmoll J, Malinchoc M, Hay JE. Fractures and avascular necrosis before and after orthotopic liver transplantation: long-term follow-up and predictive factors. Hepatology 2007;46:1198e207. [18] National Clinical Guideline C. National Institute for Health and Clinical Excellence: guidance. In: Osteoporosis: fragility fracture risk: osteoporosis: assessing the risk of fragility fracture. London: Royal College of Physicians (UK): National Clinical Guideline Centre; 2012. p. 15e48. [19] Appachi S, Kashyap SR. ‘Adiposopathy’ and cardiovascular disease: the benefits of bariatric surgery. Curr Opin Cardiol 2013; 28:540e6. [20] Luu C, Arrington AK, Falor A, Kim J, Lee B, Nelson R, et al. Impact of gastric cancer resection on body mass index. Am Surg 2014;80:1022e5. [21] Johansson H, Kanis JA, Oden A, McCloskey E, Chapurlat RD, Christiansen C, et al. A meta-analysis of the association of fracture risk and body mass index in women. J Bone Min Res 2014; 29:223e33. [22] Lalmohamed A, de Vries F, Bazelier MT, Cooper A, van Staa TP, Cooper C, et al. Risk of fracture after bariatric surgery in the United Kingdom: population based, retrospective cohort study. BMJ 2012;345:e5085. [23] Compston JE, Watts NB, Chapurlat R, Cooper C, Boonen S, Greenspan S, et al. Obesity is not protective against fracture in postmenopausal women: GLOW. Am J Med 2011;124:1043e50. [24] Black DM, Arden NK, Palermo L, Pearson J, Cummings SR. Prevalent vertebral deformities predict hip fractures and new
36
[25]
[26]
[27]
[28]
[29]
[30]
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36 vertebral deformities but not wrist fractures. J Bone Mineral Res 1999;14:821e8. Lindsay R, Silverman SL, Cooper C, Hanley DA, Barton I, Broy SB, et al. Risk of new vertebral fracture in the year following a fracture. JAMA 2001;285:320e3. Puisto V, Helio¨vaara M, Impivaara O, Jalanko T, Kro¨ger H, Knekt P, et al. Severity of vertebral fracture and risk of hip fracture: a nested caseecontrol study. Osteoporos Int 2011;22: 63e8. Sajjan SG, Barrett-Connor E, McHorney CA, Miller PD, Sen SS, Siris E. Rib fracture as a predictor of future fractures in young and older postmenopausal women: National Osteoporosis Risk Assessment (NORA). Osteoporos Int 2012;23:821e8. Barrett-Connor E, Nielson CM, Orwoll E, Bauer DC, Cauley JA. Epidemiology of rib fractures in older men: osteoporotic fractures in men (MrOS) prospective cohort study. BMJ 2010;340:c1069. Kwon SJ, Hahm JS, Cho YJ, Ahn Y, Shin DI. The influence of gastrectomy on the change of bone metabolism and bone density. Korean J Intern Med 2000;15:25e31. Nilas L, Christiansen C. Influence of PTH and 1,25(OH)2D on calcium homeostasis and bone mineral content after gastric surgery. Calcif Tissue Int 1985;37:461e6.
[31] Bisballe S, Eriksen EF, Melsen F, Mosekilde L, Sorensen OH, Hessov I. Osteopenia and osteomalacia after gastrectomy: interrelations between biochemical markers of bone remodelling, vitamin D metabolites, and bone histomorphometry. Gut 1991; 32:1303e7. [32] Maier GW, Kreis ME, Zittel TT, Becker HD. Calcium regulation and bone mass loss after total gastrectomy in pigs. Ann Surg 1997;225:181e92. [33] Baek KH, Jeon HM, Lee SS, Lim DJ, Oh KW, Lee WY, et al. Short-term changes in bone and mineral metabolism following gastrectomy in gastric cancer patients. Bone 2008;42:61e7. [34] Bernstein CN, Leslie WD, Leboff MS. AGA technical review on osteoporosis in gastrointestinal diseases. Gastroenterology 2003; 124:795e841. [35] Lucey MR, Terrault N, Ojo L, Hay JE, Neuberger J, Blumberg E, et al. Long-term management of the successful adult liver transplant: 2012 practice guideline by the American Association for the Study of Liver Diseases and the American Society of Transplantation. Liver Transplant 2013;19:3e26. [36] Suzuki Y, Fukushima S, Iwai T, Ishibashi Y, Omura N, Hanyu N, et al. Bisphosphonate incadronate prevents total gastrectomy-induced osteopenia in rats. Bone 2004;35:1346e52.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.ejcancer.com
Original Research
Fracture after gastrectomy for gastric cancer: A long-term follow-up observational study Hyun Jin Oh a, Chul-Hyun Lim a,*, Byung-Ho Yoon b, Seung Bae Yoon a, Myong Ki Baeg a, Won Chul Kim a, Yu Kyung Cho a, Jae Myung Park a, Myung-Gyu Choi a, Han Mo Yoo c, Kyo Young Song c, Hae Myung Jeon c, Cho Hyun Park c a Division of Gastroenterology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Republic of Korea b Department of Orthopaedic Surgery, Inje University College of Medicine, Seoul Paik Hospital, Republic of Korea c Division of Gastrointestinal Surgery, Department of Surgery, College of Medicine, The Catholic University of Korea, Republic of Korea
Received 26 May 2016; received in revised form 8 November 2016; accepted 16 November 2016
KEYWORDS Gastric cancer; Gastrectomy; Osteoporosis; Fracture; Risk factor
Abstract Aim: Gastrectomy is a known risk factor for decreased bone mass. We aimed to evaluate the cumulative incidence and predictive factors of fracture in gastric cancer patients who underwent gastrectomy. Method: We retrospectively reviewed the records of 1687 patients who underwent gastrectomy for gastric cancer at our hospital between September 1991 and December 2008. The exclusion criteria were stage IV gastric cancer, history of cancer recurrence, medical conditions that cause osteoporosis and high-energy injury. Fractures at sites considered to be associated with osteoporosis were diagnosed radiologically. Results: In total, our analysis included the records of 1131 patients. The incidence of postgastrectomy fracture was 42.1 cases per 1000 person-years. Fractures typically occurred within 3.7 0.5 years postoperatively. The cumulative incidence of fracture was 9.1%, 19.7%, and 37.3% by postoperative year 2, 4, and 6, respectively. During the following years, the cumulative incidence increased slowly, up to a final 40.6%. Multivariate analysis showed that older age (hazard ratio, 1.03; 95% confidence interval, 1.01e1.04) and smoking (hazard ratio, 1.35; 95% confidence interval, 1.05e1.73) were significantly associated with fracture, whereas sex, body mass index, percent weight loss, diabetes mellitus, tumour stage, and type of gastrectomy were not.
* Coresponding author: Division of Gastroenterology, Department of Internal Medicine, College of Medicine, Seoul St. Mary’s Hospital, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 137-701, Republic of Korea. Fax: þ82 2 2258 2089. E-mail address: [email protected] (C.-H. Lim). http://dx.doi.org/10.1016/j.ejca.2016.11.023 0959-8049/ª 2016 Elsevier Ltd. All rights reserved.
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
29
Conclusion: The cumulative incidence of fracture is high in gastric cancer patients who have undergone gastrectomy, and fracture rate is higher during the early postoperative period. Old age and smoking are independent risk factors for postgastrectomy fracture in these patients. More detailed postoperative surveillance and pharmacological intervention should be considered to prevent fracture. ª 2016 Elsevier Ltd. All rights reserved.
1. Introduction Gastric cancer is the 4th most common type of cancer worldwide [1], and the incidence of the disease is especially high in East Asia. Early diagnosis and curative treatment such as gastrectomy can improve the outcomes. In Korea, the survival rate for gastric cancer is over 90% when diagnosed early [2]. However, gastrectomy is a known risk factor for osteoporosis [3e5], and although the pathogenesis has not been fully elucidated, the reported incidence of osteoporosis after gastrectomy ranges from 32% to 42% [3,6,7], which implies a higher risk of fractures. Fractures decrease quality of life, and increase the economic burden on individuals, as well as on society [8e10]. In a large cohort study involving peptic ulcer patients, the incidence of osteoporotic fracture after gastrectomy was up to 70% [6]. However, to date, the incidence, characteristics, and risk factors associated with fracture after gastrectomy have not been studied longitudinally and systemically in a large cohort of gastric cancer patients. Thus, the present study aimed to evaluate the cumulative incidence and predictive factors of postgastrectomy fracture in gastric cancer patients who underwent gastrectomy.
relevant data, including clinical, surgical, and pathological records, were collected retrospectively from the database. The Institutional Review Board of our institution approved this study (KC16RISI0380). 2.2. Follow-up measurements All included patients were registered and followed-up by the Comprehensive Cancer Institute of Seoul St. Mary’s Hospital. The patients were followed-up every 3 months during the first two postoperative years, every 6 months during the next three postoperative years, and annually after the fifth postoperative year. The follow-up program consisted of interim history taking, physical examination, imaging investigations, endoscopic examination, haematology tests and blood chemistry panels. Imaging investigations included chest radiography, abdominal computerised tomography (CT) and bone scintigraphy. Clinical and laboratory parameters regarding nutritional status, including body mass index (BMI), serum albumin, and serum calcium levels were reviewed at baseline and at one year postoperatively. The percentage of weight loss was calculated as the difference between the baseline and 1-year body weight, divided by the baseline body weight.
2. Patients and methods 2.3. Outcomes measured 2.1. Patients From September 1991 to December 2008, a total of 1687 patients with gastric cancer underwent curative gastrectomy at the Seoul St. Mary’s Hospital, Seoul, Korea. Gastrectomy was defined as margin-negative surgical resection, with no residual tumour (macroscopically). The exclusion criteria were: age >70 years at gastrectomy regarded as decreasing calcium absorption (n Z 334) [11,12], postoperative tumour recurrence (n Z 118), another type of cancer (n Z 33). Among the remaining 1202 patients, those with another medical condition (n Z 26) or who underwent surgery (n Z 45) known to affect bone mineral density were also excluded. Therefore, the records of a total of 1131 patients were included in the final analysis (Fig. 1). Cancers were staged in accordance with the tumour node metastasis classification system used by the Union for International Cancer Control, 6th edition [13]. All
Fracture was diagnosed by annual bone scintigraphy, with additional CT, magnetic resonance imaging, or Xray investigations performed as clinically indicated, at the site of bone pain with or without trauma, or as needed for the differential diagnosis from bone metastasis. All radiographs, bone scans and magnetic resonance imaging scans were analysed for fractures by trained radiologists. Only those fractures at sites considered to be associated with osteoporosis (i.e. spine, coccyx, pelvis, ribs, humerus, forearm, elbow, hip, femur, tibia, fibula, clavicle, scapula, and sternum) were considered in this study. The trauma history of the patients with fractures was investigated to differentiate low-energy trauma fracture (following fall from standing height or less) from high-energy trauma fracture (following fall from a height greater than standing height). Person-years were counted from enrolment to the date of death or the end of follow-up.
30
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 1. Flow chart of enrolment and outcomes.
2.4. Statistical analysis Pearson’s chi-square test and the Student t test for unpaired data were applied for categorical and continuous variables, respectively, to compare the groups of patients with and without postgastrectomy fracture with respect to clinicopathological characteristics. A P-value of <0.05 was regarded as significant. The cumulative incidence of postgastrectomy fracture was calculated by the KaplaneMeier method, with the date of gastrectomy as the starting point. Differences in incidence were examined by the log-rank test. Multivariate analysis was performed using a Cox proportional hazards model with a backward stepwise selection procedure. All analyses were performed using SAS for Windows (version 8.02, SAS Institute, Cary, NC, USA). 3. Results 3.1. Characteristics of the study population Our study involved 1131 patients, with 344 women (30.4%), of whom 193 were postmenopausal. The mean
age was 53.8 years (range, 20e69 years). Distal subtotal gastrectomy was the most commonly performed procedure (74.0%, n Z 837). The majority of tumours (74.9%, n Z 848) were evaluated as stage I (Table 1). The patients were followed-up for 9216 person-years. The mean follow-up period was 8.4 4.2 years. Of 1131 patients, 336 (29.7%) had postgastrectomy fracture. 3.2. Incidence of postgastrectomy fracture The incidence of postgastrectomy fracture was 42.1 cases per 1000 person-years. Fractures occurred on average 3.7 0.5 years postoperatively. KaplaneMeier analysis showed a cumulative incidence of 9.1%, 19.7%, and 37.3% by postoperative year 2, 4 and 6, respectively. During the following years, the incidence of fractures continued to increase slowly, up to a final 40.6% (Fig. 2A). The incidence of postgastrectomy fracture was 48.7 and 30.1 cases per 1000 person-years for males and females, respectively, with significantly higher incidence in males (P < 0.001; Fig. 2B). The incidence of postgastrectomy fracture was also higher in
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
31
Table 1 Baseline characteristics of patients. Measures Age (Mean SD) Male Smoking Diabetes mellitus Postmenopausal state Preoperative BMI (kg/m2, Mean SD) Anaemiaa Alkaline phosphatase (U/L) Type of surgery Total gastrectomy Subtotal gastrectomy Postoperative CTx Follow-up duration (month) Mean SD Range Tumour size (mm, Mean SD) T stage T1 T2 T3 T4 N stage N0 N1 N2 N3 Pathological stage I II III Histology Differentiated Undifferentiated NA
Total population (N Z 1131)
Postgastrectomy fracture Yes (N Z 336)
No (N Z 795)
53.8 9.9 787 (69.6%) 313 (27.7%) 68 (6.0%) 193 (193/344, 56.1%) 23.3 3.1 325 (28.7%) 118.07 64.3
55.8 9.1 255 (75.9%) 115 (34.4%) 25 (7.5%) 58 (58/81, 71.6%) 23.3 3.1 91 (27.1%) 121.0 73.3
52.9 10.2 532 (66.9%) 198 (25%) 43 (5.4%) 135 (135/263, 51.3%) 23.3 3.1 234 (29,5%) 117.7 60.1
294 (26.0%) 837 (74.0%) 218 (19.3%)
83 (24.7%) 253 (75.3%) 69 (20.8%)
211 (26.5%) 584 (73.5%) 149 (18.6%)
101.0 50.2 4e279 35.1 22.8
100.8 45.3 4e279 34.4 22.3
101.1 52.1 8e261 35.4 23.0
696 (61.5%) 328 (29.0%) 104 (9.2%) 3 (0.3%)
209 (62.0%) 94 (28.0%) 31 (9.3%) 2 (0.6%)
487 (61.2%) 234 (29.5%) 73 (9.3%) 1 (0.1%)
810 (71.6%) 151 (13.4%) 102 (9.0%) 68 (6.0%)
227 (67.6%) 50 (14.9%) 32 (9.5%) 27 (8.0%)
583 (73.3%) 101 (12.7%) 70 (8.8%) 41 (5.2%)
848 (74.9%) 194 (17.2%) 89 (7.9%)
246 (73.2%) 54 (16.1%) 36 (10.7%)
602 (75.7%) 140 (17.6%) 53 (6.7%)
527 (46.7%) 601 (53.1%) 3 (0.3%)
169 (50.3%) 167 (49.7%) 0 (0%)
358 (45.0%) 434 (54.6%) 3 (0.4%)
P-value <0.001 0.003 0.001 0.189 0.001 0.904 0.418 0.426 0.519
0.581 0.929
0.499 0.189
0.202
0.067
0.226
Where appropriate, data are shown as the mean SD. Abbreviations: BMI, body mass index; CTx, chemotherapy; NA, not applicable; SD, standard deviation. a Anaemia was defined by World Health Organisation criteria: Hb < 12 g/dL in women and <13 g/dL in men.
postmenopausal women than in premenopausal women (P Z 0.002; Fig. 2C). To assess the impact of BMI on fracture incidence, each patient was assigned to one of three BMI-based groups: underweight (<18.5 kg/m2), normal 2 (18.5e24.9 kg/m ) or overweight (25.0 kg/m2) [14]. This assignment was performed once for preoperative BMI, and again for postoperative BMI. The incidence of postgastrectomy fracture was similar among the preoperative BMI-based groups, as well as among the 1year postoperative BMI-based groups. Similarly, no significant association with the incidence of postgastrectomy fracture was found regarding percent weight loss during postoperative year 1 (weight loss group, 10%; stable weight group, <10%), or type of surgery (total or subtotal gastrectomy). 3.3. Location of postgastrectomy fractures Out of the 336 patients with postgastrectomy fracture (Table 2), 90.7% had solitary fracture, whereas 9.9% had
multiple fractures. The following types of solitary fractures occurred: rib fractures, in 180 patients (59.4% of all patients with solitary fracture), within 3.8 0.2 years postoperatively; spinal fractures, in 80 patients (26.4%), within 4.2 0.4 years postoperatively; hip fracture, 12 patients (4.1%), within 2.3 0.4 years postoperatively; wrist fracture, in 10 patients (3.3%), within 3.8 0.4 years postoperatively; and other fractures (humerus, clavicle, sternum and tibia), in 21 patients (6.9%). Of the 33 patients with multiple fractures, 17 had hip fractures, whereas 11 had vertebral fracture accompanied by fracture at other sites. Multiple fractures occurred within 3.7 0.5 years postoperatively. 3.4. Risk factors for postgastrectomy fracture Compared with the group of patients without fracture, fracture group contained older patients (55.8 9.1 years versus 52.9 10.2 years; P < 0.001), more males (75.9% versus 66.9%; P Z 0.003), more smokers (34.4% versus 25%; P < 0.001), and more postmenopausal women
32
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 2. Cumulative incidence of the postgastrectomy fracture in gastric cancer patients for (A) overall, (B) male and female and (C) postmenopausal and premenopausal females.
(71.6% versus 51.3%; P < 0.001; Table 1). There were no significant differences between the groups with respect to incidence of diabetes mellitus, preoperative BMI, type of surgery, postoperative chemotherapy treatment and tumour stage. Multivariate Cox analysis showed that older age was a significant factor associated with postgastrectomy fracture (hazard ratio, 1.03; 95% confidence interval, 1.01e1.04; Table 3). Among the other variables, smoking (hazard ratio, 1.35; 95% confidence interval, 1.05e1.73) was also an independent risk factor for postgastrectomy fracture, whereas sex, BMI, percent
weight loss, diabetes mellitus, tumour stage or type of gastrectomy were not significantly associated with the incidence of fracture. 3.5. Postgastrectomy changes related to nutritional status BMI and laboratory parameters indicating nutritional status were compared between the two groups of patients (Table 4). All recorded levels of calcium and albumin were within normal limits. At baseline, albumin levels were lower in patients with postgastrectomy fracture and had increased significantly in both groups
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Table 2
Table 4
Sites of postgastrectomy fracture. Anatomic site
Number of patients
Vertebra Ribs Femur Wrist Othersa
303 (90.1%) 80 180 12 10 21 33 (9.9%)
Solitary bone fracture
Multiple bone fracture a
33
Parameters of patients with postgastrectomy fracture and without postgastrectomy fracture. With fracture
Without fracture
P
4.0 0.4 4.3 0.5a
4.1 0.3 4.3 0.3a
<0.001 0.701
8.6 0.7 8.9 0.5a
8.9 0.5 8.9 0.4
<0.001 0.384
23.3 3.1 21.0 2.8a
23.3 3.1 21.0 2.8a
0.344 0.652
b
Including humerus, clavicle, sternum and tibia.
by postoperative year 1. Similarly, at baseline, calcium levels were significantly lower in patients with fracture, but, by postoperative year 1, had increased only in these patients. A significant decrease in BMI was found at 1 year postoperatively for both groups. 3.6. Postgastrectomy fracture and overall survival Fig. 3 shows the postoperative overall survival curves of patients according to the postgastrectomy fracture. The 5-year overall survival rate was not different between two groups (98.1% versus 97.8%). During the follow-up years, overall survival rate was lower in patients with postgastrectomy fracture (76.4% versus 73.4%). However, no significant difference was observed between two groups (P Z 0.880). 4. Discussion The present study aimed to evaluate the postgastrectomy fracture risk in gastric cancer patients. Our retrospective analysis reviewed 20 years of experience at a single tertiary institution, and demonstrated high cumulative incidence of postgastrectomy fracture in such patients, as we found radiologic and historical evidence of postgastrectomy fracture in 42.1 cases per 1000 person-years. Moreover, since 18.7% of the patients were followed-up for less than 5 years postoperatively, the incidence of fractures may be underestimated. Age and smoking were independent prognostic factors after adjustment for sex, menopausal state and cancer stage,
Albumin (g/dL) Baseline 1 year after surgery Calcium (mg/dL)b Baseline 1 year after surgery BMI Baseline 1 year after surgery
Where appropriate, data are shown as the mean SD. a P < .05 (paired t test; baseline versus 1 year after surgery). b Albumin and calcium were evaluated in 465 and 460 patients, respectively.
whereas baseline and 1-year postoperative BMIs were not associated with the incidence of postgastrectomy fracture. To the best of our knowledge, this is the first long-term follow-up observational cohort study to investigate the incidence and risk factors of postgastrectomy fracture in gastric cancer patients. Between 2005 and 2008, the incidence of osteoporosis-related fractures in the general Korean population aged >50 years was 1635 cases per 100,000 person-years [15], which is over 2.5 times lower than the incidence of postgastrectomy fracture we found in the cohort of gastric cancer patients. In males, we found a much higher incidence than that reported for the general population (48.7 cases per 1000 person-years versus 725 cases per 100,000 person-years); this trend was also true for females, but to a smaller extent (30.1 cases per 1000 person-years in our cohort, versus 2408 cases per 100,000 person-years in the general population). In contrast to the other studies [15e17], we found a higher incidence of fracture in males. This is likely related to the fact that gastric cancer is more frequent in males, and our cohort was predominantly male (70%). Moreover, the proportion of smokers was higher among males than among females (38.6% versus 2.6%), and we found that smoking was independently associated with the risk for postgastrectomy fracture. Smoking is a well
Table 3 Cox proportional hazard models for cumulative incidence of postgastrectomy fracture in gastric cancer patients. Variables
Age Male versus female Postmenopausal versus premenopausal BMI Body weight change (%) Diabetes mellitus Smoking Stage III versus stage I Subtotal gastrectomy versus. total gastrectomy
Univariate analysis
Multivariate analysis
HR
95% CI
P value
HR
95% CI
P value
1.03 1.58 2.16 1.01 1.00 1.30 1.43 1.51 1.08
1.02e1.04 1.23e2.03 1.33e3.50 0.97e1.04 0.99e1.01 0.87e1.96 1.14e1.79 1.06e2.14 0.84e1.39
<0.001 <0.001 0.002 0.786 0.929 0.202 0.002 0.021 0.528
1.03 1.15 1.54
1.01e1.04 0.35e3.76 0.92e2.57
<0.001 0.820 0.095
1.35 1.41
1.05e1.73 0.99e2.00
0.018 0.055
Abbreviations: HR, hazard ratio; CI, confidence interval; BMI, body mass index.
34
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
Fig. 3. Overall survival according to the postgastrectomy fracture.
known risk factor for fracture in general population [16]. The Fracture Risk Assessment Tool is used in clinical settings to predict the probability of major fracture based on clinical risk factors such as BMI [18]. Body weight loss of 5e15% typically occurs after gastrectomy, and is maintained without significant changes until 1 year postoperatively [19,20]. In our study, the percent body weight loss was within these limits. In general, body weight loss or lower BMI have been known to increase the risk of fracture, but this association has recently been challenged. Our study suggested that percentage of body weight loss, baseline BMI, and 1year postoperative BMI are not associated with the incidence of fractures. Our results are consistent with the findings presented in recent reports for general population as well as patients with bariatric surgery [16,21e23]. In the patients included in the present study, the most common fracture sites were the ribs and the spine, which is similar to reports of fractures after liver transplantation [17]. In addition, our study included the incident osteoporotic fractures. In general, these asymptomatic fractures are relevant because of increasing risk of subsequent osteoporotic fractures [24e26]. In other large-population studies, such as the National Osteoporosis Risk Assessment involving postmenopausal women, or the Osteoporotic Fractures in Men study, rib fracture was the most common and important incident clinical fracture, and it was found that a history of rib fracture carried at least a 2-fold increased risk of an incident subsequent osteoporotic fracture [27,28]. The pathogenesis of bone loss and fracture in gastric cancer patients who underwent gastrectomy is uncertain. One possible explanation is malabsorption. Specifically, calcium is absorbed primarily in the duodenum, and gastrectomy with anastomosis alters normal gastrointestinal physiology. Gastric dumping and the formation of insoluble calcium soaps due to
malabsorption of fat may result in malabsorption of calcium. Decreased acid secretion in the gastric mucosa also may result in malabsorption of calcium. However, serum calcium levels of postgastrectomy patients were reported to be within normal limits [4,29,30], consistent with our data. Moreover, 25-hydroxyvitamin D levels are within normal or, at most, subnormal levels [30,31]. Calcium homoeostasis and elevated parathyroid (PTH) hormones after gastrectomy may explain the normal calcium levels and bone loss after gastrectomy. The mean serum PTH levels increase after gastrectomy [32,33], and patients with higher postoperative PTH level exhibited a more pronounced decrease in bone mass density (BMD). In our study, patients with postgastrectomy fracture exhibited a significant increase in serum calcium levels by 1 year postoperatively, which may reflect maintenance of serum calcium homoeostasis at the expense of BMD, subsequently leading to higher risk of fractures. Thus, a remodelling imbalance with dominant bone resorption appears to contribute to bone loss and fracture in these patients. There are no specific screening guidelines and no established risk factor grading system for bone health in gastric cancer patients. Although, based on insufficient data, American Gastroenterological Association recommended BMD measurement in gastrectomy patients [34], best surveillance schedule and method were not defined. For liver transplantation patients, during the first 5 postoperative years, BMD should be measured annually in osteopenic patients, and every 2e3 years in patients with normal BMD [35]. We found that fracture mostly occurred within the first 6 postoperative years, but the cumulative incidence continued to increase afterwards. In a prospective study involving gastrectomy patients, bone remodelling imbalance occurred early within the first postoperative year [33]. Therefore, bone pain or fracturing, serum calcium, phosphorous, 25hydroxyvitamin D and PTH levels should be assessed at least within 6 years postoperatively. Regular follow-
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36
up tests and appropriate intervention should also be considered. Dominant bone resorption due to calcium homoeostasis and secondary hyperparathyroidism may also occur in the early postgastrectomy period, in which case dietary calcium and bisphosphonate can be administered prophylactically. In rats, bisphosphonate was effective in blocking gastrectomy-induced decreases in BMD of the lumbar spine and femoral bone, and in femoral bending strength [36]. Although comparable human studies are scarce, we believe bisphosphonate may be used to prevent bone resorption and fracture in gastric cancer patients who underwent gastrectomy, along with dietary calcium and vitamin D supplements even when serum calcium levels are within normal limits. The results of the present study are highly relevant because: the incidence and location of fractures was assessed based on radiologic reports and low trauma history; the analysis was longitudinal; the study population was large; we also investigated parameters particularly important in the target population, namely baseline and postoperative BMIs, as well as nutritionrelated laboratory parameters. This study is limited due to its retrospective nature, and the fact that we could not obtain serum markers for bone metabolism, 25hydroxyvitamin D, PTH, or BMD data. Although our study focused on the fracture risk of gastric cancer patients who underwent curative gastrectomy, it can be extended to the advanced gastric cancer in future study. In conclusion, the cumulative incidence of fracture is high in patients with gastric cancer undergone gastrectomy, and high rate of fracturing is reported at the early period of 6 years after gastrectomy. Old age and smoking are the independent risk factors for postgastrectomy fracture. More intensive surveillance pharmacologic interventions should be considered to prevent fracture. Funding and support This research did not receive any specific grant from funding agencies in the public, commercial or not-forprofit sectors. Conflict of interest statement None declared.
References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69e90. [2] Lee H-J, Yang H-K, Ahn Y-O. Gastric cancer in Korea. Gastric Cancer 2002;5:0177e82.
35
[3] Inoue K, Shiomi K, Higashide S, Kan N, Nio Y, Tobe T, et al. Metabolic bone disease following gastrectomy: assessment by dual energy X-ray absorptiometry. Br J Surg 1992;79:321e4. [4] Liedman B, Bosaeus I, Mellstrom D, Lundell L. Osteoporosis after total gastrectomy. Results of a prospective, clinical study. Scand J Gastroenterol 1997;32:1090e5. [5] Tovey FI, Hall ML, Ell PJ, Hobsley M. A review of postgastrectomy bone disease. J Gastroenterol Hepatol 1992;7: 639e45. [6] Pryor JP, O’Shea MJ, Brooks PL, Datar GK. The long-term metabolic consequences of partial gastrectomy. Am J Med 1971; 51:5e10. [7] Lim JS, Kim SB, Bang HY, Cheon GJ, Lee JI. High prevalence of osteoporosis in patients with gastric adenocarcinoma following gastrectomy. World J Gastroenterol 2007;13:6492e7. [8] Johnell O. The socioeconomic burden of fractures: today and in the 21st century. Am J Med 1997;103. 20Se5S; discussion 5Se6S. [9] Burge R, Dawson-Hughes B, Solomon DH, Wong JB, King A, Tosteson A. Incidence and economic burden of osteoporosisrelated fractures in the United States, 2005e2025. J Bone Min Res 2007;22:465e75. [10] Si L, Winzenberg TM, de Graaff B, Palmer AJ. A systematic review and meta-analysis of utility-based quality of life for osteoporosis-related conditions. Osteoporos Int 2014;25:1987e97. [11] Bullamore J, Wilkinson R, Gallagher J, Nordin B, Marshall D. Effect of age on calcium absorption. Lancet 1970;296:535e7. [12] Riggs BL. Overview of osteoporosis. West J Med 1991;154:63. [13] Sobin LH, Fleming ID. TNM classification of malignant tumors. Cancer 1997;80:1803e4. [14] WHO EC. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157. [15] Park C, Ha YC, Jang S, Jang S, Yoon HK, Lee YK. The incidence and residual lifetime risk of osteoporosis-related fractures in Korea. J Bone Min Metab 2011;29:744e51. [16] Crandall CJ, Yildiz VO, Wactawski-Wende J, Johnson KC, Chen Z, Going SB, et al. Postmenopausal weight change and incidence of fracture: post hoc findings from women’s health initiative observational study and clinical trials. BMJ 2015;350: h25. [17] Guichelaar MM, Schmoll J, Malinchoc M, Hay JE. Fractures and avascular necrosis before and after orthotopic liver transplantation: long-term follow-up and predictive factors. Hepatology 2007;46:1198e207. [18] National Clinical Guideline C. National Institute for Health and Clinical Excellence: guidance. In: Osteoporosis: fragility fracture risk: osteoporosis: assessing the risk of fragility fracture. London: Royal College of Physicians (UK): National Clinical Guideline Centre; 2012. p. 15e48. [19] Appachi S, Kashyap SR. ‘Adiposopathy’ and cardiovascular disease: the benefits of bariatric surgery. Curr Opin Cardiol 2013; 28:540e6. [20] Luu C, Arrington AK, Falor A, Kim J, Lee B, Nelson R, et al. Impact of gastric cancer resection on body mass index. Am Surg 2014;80:1022e5. [21] Johansson H, Kanis JA, Oden A, McCloskey E, Chapurlat RD, Christiansen C, et al. A meta-analysis of the association of fracture risk and body mass index in women. J Bone Min Res 2014; 29:223e33. [22] Lalmohamed A, de Vries F, Bazelier MT, Cooper A, van Staa TP, Cooper C, et al. Risk of fracture after bariatric surgery in the United Kingdom: population based, retrospective cohort study. BMJ 2012;345:e5085. [23] Compston JE, Watts NB, Chapurlat R, Cooper C, Boonen S, Greenspan S, et al. Obesity is not protective against fracture in postmenopausal women: GLOW. Am J Med 2011;124:1043e50. [24] Black DM, Arden NK, Palermo L, Pearson J, Cummings SR. Prevalent vertebral deformities predict hip fractures and new
36
[25]
[26]
[27]
[28]
[29]
[30]
H.J. Oh et al. / European Journal of Cancer 72 (2017) 28e36 vertebral deformities but not wrist fractures. J Bone Mineral Res 1999;14:821e8. Lindsay R, Silverman SL, Cooper C, Hanley DA, Barton I, Broy SB, et al. Risk of new vertebral fracture in the year following a fracture. JAMA 2001;285:320e3. Puisto V, Helio¨vaara M, Impivaara O, Jalanko T, Kro¨ger H, Knekt P, et al. Severity of vertebral fracture and risk of hip fracture: a nested caseecontrol study. Osteoporos Int 2011;22: 63e8. Sajjan SG, Barrett-Connor E, McHorney CA, Miller PD, Sen SS, Siris E. Rib fracture as a predictor of future fractures in young and older postmenopausal women: National Osteoporosis Risk Assessment (NORA). Osteoporos Int 2012;23:821e8. Barrett-Connor E, Nielson CM, Orwoll E, Bauer DC, Cauley JA. Epidemiology of rib fractures in older men: osteoporotic fractures in men (MrOS) prospective cohort study. BMJ 2010;340:c1069. Kwon SJ, Hahm JS, Cho YJ, Ahn Y, Shin DI. The influence of gastrectomy on the change of bone metabolism and bone density. Korean J Intern Med 2000;15:25e31. Nilas L, Christiansen C. Influence of PTH and 1,25(OH)2D on calcium homeostasis and bone mineral content after gastric surgery. Calcif Tissue Int 1985;37:461e6.
[31] Bisballe S, Eriksen EF, Melsen F, Mosekilde L, Sorensen OH, Hessov I. Osteopenia and osteomalacia after gastrectomy: interrelations between biochemical markers of bone remodelling, vitamin D metabolites, and bone histomorphometry. Gut 1991; 32:1303e7. [32] Maier GW, Kreis ME, Zittel TT, Becker HD. Calcium regulation and bone mass loss after total gastrectomy in pigs. Ann Surg 1997;225:181e92. [33] Baek KH, Jeon HM, Lee SS, Lim DJ, Oh KW, Lee WY, et al. Short-term changes in bone and mineral metabolism following gastrectomy in gastric cancer patients. Bone 2008;42:61e7. [34] Bernstein CN, Leslie WD, Leboff MS. AGA technical review on osteoporosis in gastrointestinal diseases. Gastroenterology 2003; 124:795e841. [35] Lucey MR, Terrault N, Ojo L, Hay JE, Neuberger J, Blumberg E, et al. Long-term management of the successful adult liver transplant: 2012 practice guideline by the American Association for the Study of Liver Diseases and the American Society of Transplantation. Liver Transplant 2013;19:3e26. [36] Suzuki Y, Fukushima S, Iwai T, Ishibashi Y, Omura N, Hanyu N, et al. Bisphosphonate incadronate prevents total gastrectomy-induced osteopenia in rats. Bone 2004;35:1346e52.