Expression of Concern Expression of Concern on “ Influence of gastrectomy on cortical and cancellous bones in rats

Gastroenterology Research and Practice would like to express our concern with the article titled “Influence of gastrectomy on cortical and cancellous bones in rats”, published in 2013 [1]. It was brought to our attention that a later article by the same authors published in Journal of Nutritional Science and Vitaminology in 2014 [2] used some of the same animals as in the Gastroenterology Research and Practice article, but the second article had an additional arm of rats given vitamin K and did not cite the first article. The authors did not mention in the Gastroenterology Research and Practice article that they had also used an additional arm of rats given vitamin K. In addition to the partial redundant publication, readers should be aware of the following concerns and discrepancies:

Gastroenterology Research and Practice would like to express our concern with the article titled "Influence of gastrectomy on cortical and cancellous bones in rats", published in 2013 [1].It was brought to our attention that a later article by the same authors published in Journal of Nutritional Science and Vitaminology in 2014 [2] used some of the same animals as in the Gastroenterology Research and Practice article, but the second article had an additional arm of rats given vitamin K and did not cite the first article.The authors did not mention in the Gastroenterology Research and Practice article that they had also used an additional arm of rats given vitamin K.In addition to the partial redundant publication, readers should be aware of the following concerns and discrepancies: (1) In Figure 1 of Journal of Nutritional Science and Vitaminology, the gastrectomy data for all six variables are discrepant with those for the same variables in Figure 3 of Gastroenterology Research and Practice (2) In Figure 1 of Journal of Nutritional Science and Vitaminology, the control data for work to failure in the femoral diaphysis are discrepant with those for the same variable in Figure 3 of Gastroenterology Research and Practice (3) In the legends to Figures 1, 2, and 3 in Gastroenterology Research and Practice, the superscript letters denoting statistical significance are stated to be related to 'versus gastrectomy' though they are above the gastrectomy data.This should probably have read 'versus control'; (4) In Journal of Nutritional Science and Vitaminology, control and gastrectomy groups had vehicle administered by oral gavage, but in Gastroenterology Research and Practice there is no mention of vehicle administration in either group (5) No funding statement was provided (6) The study was stated to have been conducted at Hamri Co (Ibaraki, Japan), yet none of the authors had affiliations there and neither did any of the acknowledged contributors (7) Co-author Yoshihiro Sato, since deceased, previously admitted he was given gift authorship on other articles by Jun Iwamoto We asked the first author, Jun Iwamoto, to respond to these concerns and provide the raw data, but he did not explain the discrepancies or provide the data.We asked the Dean of Keio University School of Medicine and the Provost of Keio University to investigate.We were informed by a contact from the Research Administration at Keio University School of Medicine that an investigation was complete and we would be contacted by Dr. Iwamoto, but we were not told the outcome of the investigation and our specific concerns were not addressed.
Dr. Iwamoto provided the raw data for both articles (sending one set of spreadsheets with data comparing controls versus gastrectomy, and other set of spreadsheets comparing controls versus gastrectomy versus gastrectomy plus vitamin K, all available as supplementary files (available here)), but he did not explain which data were associated with which of the results in the articles.The control results and data are the same in the two articles and spreadsheets.However, the gastrectomy results differ between the spreadsheets, because not all the same animals were reported in

Introduction
Gastric surgery is mostly needed for the treatment of gastric cancer.After gastrectomy (GX), the daily dietary intake of calcium, magnesium, phosphorus, iron, zinc, vitamin D, vitamin B 12 , and folic acid is reported to be lower than the recommendations [1][2][3].In particular, GX impairs calcium and vitamin D metabolism, leading to a risk of bone disease including not only osteoporosis, but also osteomalacia or a mixed pattern of osteoporosis osteomalacia with secondary hyperparathyroidism [4][5][6][7].The risk for vertebral and hip fractures is increased in GX patients [7][8][9][10][11][12].However, a strategy for preventing fractures in GX patients has not yet been established.Preclinical studies using animals could be useful for identifying available interventions to prevent fractures in GX patients.
An animal model of osteopenia can be created by performing GX in rats.Total GX and resection of the acid-producing part of the stomach (fundectomy) induces gastrinemia and malnutrition [13,14], thereby initiating osteopenia in rats [15][16][17][18][19].However, although bone histomorphometry studies have demonstrated GX-induced decreases in cancellous bone volume per tissue volume and cortical bone area in rats [17,18], the changes in bone formation, resorption, and osteoid parameters after GX have not been adequately studied.The difference in the responses of cortical and cancellous bones to GX remains to be established in rats.Understanding the differential influence of GX on cortical and cancellous bones is important for clarifying the effect of interventions on the skeleton in GX rats.The aim of the present study was to examine the influence of GX on cortical and cancellous bones in rats.

Handling of Animals.
Twenty male Sprague-Dawley rats (11-week old) were purchased from Charles River Japan (Kanagawa, Japan).The animals were fed a standard pellet diet containing 1.25% calcium (calcium carbonate) and 0.9% Gastroenterology Research and Practice phosphorus (CRF-1; Oriental Yeast, Co., Ltd., Tokyo, Japan).The rats were housed in a local animal room at a temperature of 24 ∘ C, a humidity of 50%, and a 12 h on/off cycle for lighting.Free access to water and a pellet diet were allowed.After 1 week of adaptation to this environment, the rats (12week old) were sorted into strata according to body weight and were then randomized using the stratified weight method into the following two groups of 10 animals each: a sham operation control (CON) group and a GX group.A total GX was performed under general anesthesia using 25-30 mg/kg of pentobarbital (Kyoritsu Seiyaku Co., Ltd., Tokyo, Japan) injected intraperitoneally together with 2%-3% isoflurane (Mylan Inc., Tokyo, Japan) delivered via a Table Top Laboratory Animal Anesthesia System (V1 Type VetEquip, Inc., CA, USA).A longitudinal incision was made on the abdomen to expose the stomach.The whole stomach was excised, and an anastomosis of the esophagus and duodenum was performed.The weight of the rats was monitored weekly, and the duration of observation was 7 weeks.This experiment was performed at the laboratory of Hamri Co., Ltd.(Ibaraki, Japan), which has been approved by the Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC) International.The experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Hamri Co., Ltd.(Ibaraki, Japan).

Preparation of Specimens.
Rats were labeled by subcutaneous injection of 10 mg/kg of calcein (Dojindo Laboratories, Kumamoto, Japan) at 7 and 3 days before death.Seven weeks after surgery, the animals were sacrificed by exsanguination under general anesthesia with 2%-3% isoflurane (Mylan Inc., Tokyo, Japan) using a Table Top Laboratory Animal Anesthesia System (V1 Type VetEquip, Inc., CA, USA).The bilateral femora and the right tibia were harvested from each animal.The length of the left femur was measured using a dial caliper, and the weight was measured using an electronic balance (A&D Company, Tokyo, Japan).Then, the femur was preserved in saline, stored in a freezer (−60 ∘ C), and processed for the mechanical strength test.The right femur and tibia were preserved in 70% ethanol and were processed for peripheral quantitative computed tomography (pQCT) and the bone histomorphometric analysis, respectively.

pQCT of the Femur.
The distal metaphysis and diaphysis of the femur were scanned using pQCT (XCT-Research SA+; Stratec Medizintechnik GmbH, Pforzheim, Germany) in 70% ethanol/saline.The bones were placed horizontally in a polypropylene tube and were scanned at a voxel size of 0.12 mm.The scan line was adjusted using the scout view, and sites 3 mm and 14 mm proximal to the distal growth plate were scanned.For the analysis, a threshold of 395 mg/cm 3 in contour mode 2 was used to separate the bone tissue from the marrow in the distal metaphysis.The total bone mineral content (BMC) and bone mineral density (BMD) were evaluated.A constant threshold of 690 mg/cm 3 in contour mode 1 was used to separate the cortical bone from the trabecular bone in the diaphysis.The cortical thickness, BMC, and BMD were evaluated.

Bone Histomorphometry of the Tibia.
The tibia was cut into two parts (proximal metaphysis and diaphysis plus distal metaphysis) with a diamond band saw (EXAKT BS 3000; Norderstedt, Germany), and each part was stained according to the method of Villanueva [20].After dehydration with ethanol and acetone, the bone tissue was embedded in methyl methacrylate (Wako Pure Chemical, Japan).Frontal sections of the tibial proximal metaphysis were cut at a thickness of 5 m on a microtome (Leica RM 2065; Nussloch, Germany).Cross-sections of the tibial diaphysis were obtained 4 mm proximal to the tibiofibular junction at a thickness of 20-30 m with a microgrinder (Exakt KG 4000; Norderstedt, Germany).Then, the specimens were observed under a fluorescence microscope (Zeiss Axioplan 2; Jena, Germany) coupled with a video camera (CCD Color Camera CS 5270 I; Tokyo Electronic Industry Co., Ltd., Tokyo, Japan).A bone morphometry software program (Winroof Version 3.5; Mitani Corporation, Japan) was used for the histomorphometric analysis.
The parameters assessed for the diaphysis were the total tissue area (Tt.Ar), marrow area (Ma.Ar), periosteal and endocortical BS (perimeter), sLS, dLS, interlabel width, and endocortical ES, in accordance with the method described by Chen et al. [23].The endocortical OS, Oc.S, and N.Oc were also measured.Two large cavities with and without vessels, which were independent of osteocytes and lacunae [24,25], were observed in the intracortical area of each specimen to measure the cavity area (porotic Ar).These data were then used to calculate the cortical area (Ct.Ar), the periosteal and endocortical MS/BS, MAR, and BFR/BS, and the endocortical OS/BS, ES/BS, Oc.S/BS, N.Oc/BS, and Aj.AR.equilibration.The mechanical strength of the femoral diaphysis was evaluated using the three-point bending test.A load was applied to the bone midway between two supports placed 18 mm apart.The femur was positioned so that the loading point was at the center of the femoral diaphysis, and bending occurred at the medial-lateral axis.Load-displacement curves were recorded at a crosshead speed of 5 mm/min using a material-testing machine (MZ-500S; Maruto Instrument, Tokyo, Japan).The parameters analyzed were the stiffness, work to failure, and ultimate force.Just after the three-point bending test of the femoral diaphysis, the distal metaphysis of the femur was isolated over a length of 10 mm from the joint surface of the femoral condyle.The mechanical strength of this segment was then measured using a compression test.A compressive load was applied using a rectangular parallelepiped crosshead (length, 2 cm; width, 2 cm; and height, 1 cm) on the femoral distal metaphysis from the lateral aspect to the medial aspect.The specimens were positioned so that the loading point was at the center of the femoral lateral condyle.Load-displacement curves were recorded at a crosshead speed of 5 mm/min and a compression depth of 2.5 mm using a materials-testing machine (MZ500D; Maruto).The parameters analyzed were the stiffness, work to failure, and ultimate force.

Statistical Analysis.
Data were expressed as the mean ± standard deviation (SD).Comparisons among the groups were performed using an unpaired t-test.All the statistical analyses were performed using the Stat View J-5.0 program on a Windows computer, and a P value < 0.05 was regarded as significant.

Body Weight, Femoral Length, Femoral Weight, and
Femoral BMC and BMD.Table 1 shows the body weight, femoral length, femoral weight, and BMC and BMD.The initial body weight did not differ significantly between the two groups.The final body weight was lower in the GX group than in the CON group.The GX did not significantly influence the femoral length.However, the GX induced decreases in the femoral weight and the BMC and BMD of the femoral diaphysis and distal metaphysis.The GX also induced a decrease in the cortical thickness of the femoral diaphysis.The mean GX-induced reductions in the BMC and BMD of the femoral diaphysis were 22.9% and 3.2%, respectively.The mean GX-induced reductions in the BMC and BMD of the femoral distal metaphysis were 33.3% and 32.4%, respectively.2 show the results of the bone histomorphometric analysis of the tibial proximal metaphysis.The GX induced decreases in the BV/TV and Tb.N and an increase in the Tb.Sp without significantly influencing the Tb.Th (Figure 1).The mean GX-induced reduction in the BV/TV was 48.6%.These changes were associated with increased bone formation (Ob.S/BS, MS/BS, MAR, and BFR/BS) and increased O.Th, OS/BS, OV/BV, and Mlt (Table 2).However, the GX did not significantly influence bone resorption (ES/BS, Oc.S/BS, and N.Oc/BS) or the Omt (Table 2).

Bone Histomorphometric Analysis of the Tibia
Figure 2 and Table 3 show the results of the bone histomorphometric analysis of the tibial diaphysis.GX induced a decrease in the Ct.Ar and increases in the Ma.Ar and endocortical BS (Figure 2).The mean GX-induced reduction in the Ct.Ar was 16.8%.These changes were associated with increased endocortical bone resorption (ES/BS, Oc.S/BS, and N.Oc/BS) without any significant changes in endocortical bone formation (MS/BS, MAR, and BFR/BS) (Table 3).The GX did not significantly influence the Tt.Ar or the periosteal BS (Figure 2), despite the significant but modest decrease in periosteal bone formation (MAR and BFR/BS) (Table 3).The GX did not significantly influence porotic Ar or endocortical OS/BS (Table 3).

Mechanical Strength of the Femur.
Figure 3 shows the results of the mechanical strength test of the femur.In the femoral diaphysis, the GX induced a decrease in the ultimate force but did not cause any significant changes in the stiffness or work to failure.In the femoral distal metaphysis, the GX induced decreases in the stiffness, work to failure, and ultimate force.The mean GX-induced reduction in the ultimate force was 21.4% for the femoral diaphysis and 36.2% for the femoral distal metaphysis.

Discussion
The present study confirmed that GX induced cortical and cancellous osteopenia, leading to a deterioration in the bone strength of the metaphysis and diaphysis in the long bones of rats.The response of bone formation, resorption, and osteoid parameters to GX and the degree of GX-induced osteopenia and the deterioration of bone strength appeared to differ between cortical and cancellous bones.These results revealed the differential responses of cortical and cancellous bones to GX in rats.GX in rats has been reported to induce osteopenia [14,[17][18][19][26][27][28][29][30].In the present study, GX induced decreases in the BMC, BMD, ultimate force, work to failure, and stiffness of the femoral distal metaphysis, the BMC, BMD, and ultimate force of the femoral diaphysis, and the cancelous BV/TV and Ct.Ar in the tibia.The mean GX-induced reductions in the BMC, BMD, and ultimate force were greater in the femoral distal metaphysis (33.3%, 32.4%, and 36.2%,resp.)than in the femoral diaphysis (22.9%, 3.2%, and 21.4% resp.).The mean GX-induced reduction in BV/TV was also greater than that of Ct.Ar (48.6% and 16.8%, resp.).Thus, the GXinduced osteopenia and deterioration in bone strength were more severe at skeletal sites rich in cancellous bone, compared with those rich in cortical bone.In rats, cancellous osteopenia after GX appears to be similar to that which occurs after ovariectomy although the cortical osteopenia that occurs after GX may be more severe than that occurring after ovariectomy [17][18][19].

E X P R E S S I O N O F C O N C E R N
Although bone histomorphometry studies have demonstrated GX-induced decreases in cancellous BV/TV and Ct.Ar in rats [17,18], the changes in bone formation, resorption, and osteoid parameters after GX have not been adequately studied.Furthermore, cortical bone has received less attention than cancellous bone in bone histomorphometry studies.In the present study, the GX-induced decrease in cancellous BV/TV was characterized by increases in O.Th, OS/BS, OV/BV, and bone formation, without any significant alteration in bone resorption.No significant influence on Tb.Th or Omt was seen, consistent with the results of a previous study showing that GX-induced cancellous osteopenia was characterized by a normal width but a decreased maturation time of the osteoid and an increased bone formation rate in rats [14].As to cortical bone, a GX-induced decrease in Ct.Ar was characterized by increased endocortical bone resorption without any significant alterations in endocortical bone formation and OS/BS.The Tt.Ar and periosteal BS did not change significantly, consistent with the results of previous studies [28,31].These results suggested that the response of bone formation, resorption, and osteoid parameters to GX appeared to differ between cortical and cancellous bones.The risk of osteomalacia in cancellous bone must be considered after GX.GX might not influence the longitudinal bone

E X P R E S S I O N O F C O N C E R N
Gastroenterology Research and Practice 5  growth [30], resulting in the nonsignificant influence of GX on the femoral length.Gastric acid is thought to facilitate the intestinal absorption of ingested calcium by mobilizing calcium from insoluble complexes in the diet [15].After a GX in humans, pigs, and rats, the serum gastrin level decreases, calcium absorption is impaired, the serum calcium and 25(OH)D levels decrease, and the 1,25(OH) 2 D level increases [6,7,32,33].However, we did not evaluate the serum and urinary calcium and phosphorus levels, the serum 25(OH)D, 1,25(OH) 2 D, and parathyroid hormone (PTH) levels, or bone turnover markers.The influence of GX on these biochemical parameters may be important for translating the results of our preclinical study into clinical practice.Thus, further studies are needed to address this issue.
Approximately 80-90% of BMC is comprised of calcium and phosphorus.Other dietary components, such as protein, magnesium, zinc, copper, iron, fluoride, vitamins D, A, C, and K are required for normal bone metabolism [34].After GX, not only calcium, phosphorus, and vitamin D, but also magnesium, iron, zinc, vitamin B 12 , and folic acid become deficient [31][32][33].Zinc is essential for normal development of the skeleton [35].Magnesium, vitamin B 12 , and folic acid are related to bone quality [36,37].These dietary components could have played a role in the impairment of bone mass and quality in terms of bone strength in GX rats.Although gastric acid secretion and gastric acidity have been suggested to play an important role in the intestinal absorption of calcium from ingested food or calcium salts such as calcium carbonate, available evidence suggests that gastric acid secretion and gastric acidity do not normally play a role in the absorption of dietary calcium carbonate [38].Further studies are needed to clarify the contribution of malabsorption of nutrients, trace elements, and other vitamins in GX-induced changes in bone parameters in rats.Therapy of GX-induced osteopenia may consist of improving impaired calcification indicated by increased osteoid parameters in cancellous bone and increasing periosteal bone formation as well as decreasing endocortical bone resorption in cortical bone.Primarily, calcium and vitamin D supplementations are required because GX impairs calcium and vitamin D metabolism.Clinical studies showed the effects of alendronate and teriparatide on BMD in patients with postgastrectomy osteoporosis [39][40][41].Preclinical study demonstrated the effect of incadronate, alendronate, estrogen, and PTH on cancellous bone mass in GX rats [42,43].However, a strategy for the prevention of fractures in GX patient population remains to be established although bisphosphonates and teriparatide may be good candidates for therapy of GX-induced osteopenia based on the clinical data.Further studies are needed to clarify the best therapy for GXinduced osteopenia.

Conclusions
The present study compared the influence of GX on cortical and cancellous bones in rats and found that the responses of bone formation, resorption, and osteoid parameters to GX appear to differ between cortical and cancellous bones.GXinduced reductions in BMC, BMD, and bone strength were greater at skeletal sites rich in cancellous bone than at sites rich in cortical bone, and the GX-induced decrease in bone mass was greater in cancellous bone than in cortical bone.These differential responses of cortical and cancellous bones should be taken into consideration when clarifying the effect of interventions on the skeleton in GX rats.

E X P R E S S I O N O F C O N C E R N
Gastroenterology Research and Practice Test of the Femur.Each specimen was submerged for 1 hour before testing to allow temperature E X P R E S S I O N O F C O N C E R N Gastroenterology Research and Practice 3

Figure 1 :
Figure1: Bone histomorphometric analysis of the tibial proximal metaphysis.Structural parameters of cancellous bone.Data are expressed as the mean ± SD.An unpaired t-test was used to compare data between the two groups.a  < 0.05, b  < 0.001, and c  < 0.0001 versus GX.CON: control, GX: gastrectomy.

Figure 2 :Figure 3 :
Figure2: Bone histomorphometric analysis of the tibial diaphysis.Structural parameters of cortical bone.Data are expressed as the mean ± SD.An unpaired t-test was used to compare data between the two groups.a  < 0.001 and b  < 0.0001 versus GX.CON: control, GX: gastrectomy.

Table 1 :
Body weight and femoral length, weight, and BMC and BMD.
Data are expressed as the mean ± SD.An unpaired t-test was used to compare data between the two groups.GX: gastrectomy, NS: not significant.

Table 2 :
Bone histomorphometric analysis of the tibial proximal metaphysis (formative and resorptive parameters of cancellous bone).Data are expressed as the mean ± SD.An unpaired t-test was used to compare data between the two groups.

Table 3 :
Bone histomorphometric analysis of the tibial diaphysis (formative and resorptive parameters of cortical bone).Data are expressed as the mean ± SD.An unpaired t-test was used to compare data between the two groups.GX: gastrectomy, NS: not significant.