Journal of Bone and Mineral Metabolism

, Volume 28, Issue 4, pp 456–467 | Cite as

Cortical and trabecular bone microarchitecture and turnover in alcohol-induced chronic pancreatitis: a histomorphometric study

  • Christine M. Schnitzler
  • Julia M. Mesquita
  • Roy Shires
Original Article

Abstract

Alcohol-induced chronic pancreatitis is associated with bone loss, but bone histomorphometric data describing the mechanism of cortical (Ct) and trabecular (Tb) bone loss are scarce. In this case-control study, we investigated 13 black male patients aged 41.2 ± 8.9 years with alcohol-induced chronic pancreatitis by routine iliac crest cortical and trabecular histomorphometry and by biochemistry relevant to bone, liver function, and iron overload. Patients showed lower values for Ct thickness (P = 0.018), endocortical (Ec) wall thickness (P = 0.0002), Tb bone volume (0.019), Tb thickness (0.001), Tb wall thickness (P < 0.0001), Ec osteoid thickness (P = 0.001), Ec mineral apposition rate (P = 0.011), and Ec bone formation rate (P = 0.035). Ec eroded surface (P = 0.004) was elevated compared to controls. Tb osteoid thickness (P = 0.14) and Tb mineral apposition rate (P = 0.195) tended to be lower than in controls. Levels of 25-hydroxyvitamin D (P < 0.005), serum magnesium (P = 0.02), and ascorbic acid (P = 0.049) were lower and urine calcium/creatinine ratios higher than in controls. Alkaline phosphatase and gamma-glutamyl transpeptidase (GGT) were negatively correlated but iron markers were positively correlated with bone structural and formation variables. The histomorphometric data were found to be consistent with alcohol bone disease. Osteomalacia was not a feature. Secondary pathogenetic factors were liver disease, hypovitaminosis D and C, diabetes mellitus, and possibly chronic pancreatitis.

Keywords

Alcohol Chronic pancreatitis Cortical bone Trabecular bone Histomorphometry 

Introduction

Chronic pancreatitis is a progressive inflammatory condition that leads to irreversible destruction of the pancreas, resulting in exocrine and endocrine insufficiency, ultimately leading to malabsorption and diabetes mellitus [1]. In adults by far the most common cause is prolonged heavy alcohol abuse, both in South Africa [2] and elsewhere [3]. The condition has been shown to lead to bone loss and reduced levels of vitamin D metabolites, whether associated with alcohol abuse [4] or not [1]. The bone loss in alcohol-induced chronic pancreatitis is multifactorial in origin: alcohol abuse [5, 6] with its secondary abnormalities [7] and chronic pancreatitis itself with its complications [1] all have adverse effects on bone. Although the bone disease of alcohol abuse has been well characterized histomorphometrically [5, 6, 8, 9], bone histomorphometric data in alcohol-induced pancreatitis are scant. One East African autopsy study of 20 cases of “chronic pancreatic disease” reported reduced trabecular bone volume without evidence of osteomalacia but made no reference to possible causation [10].

The aim of this study was to examine cortical and trabecular iliac crest bone in patients with alcohol-induced chronic pancreatitis by histomorphometry for microarchitectural and bone turnover abnormalities to highlight the adverse effects of this condition on bone.

Materials and methods

Subjects

In this case-control study, we investigated 13 black male patients aged 41.2 ± 8.9 years (mean ± SD) (median, 37 years; range, 27–55 years) suffering from alcohol-induced chronic pancreatitis with pancreatic insufficiency by cortical and trabecular iliac crest histomorphometry and biochemistry. Six of the 13 patients had been diabetics for 1 month to 5 years, all requiring insulin. Glycemic control at the time of investigation was inadequate, as evidenced by HbA1C levels of 11.1% and 34% (normal, <6.5%). All patients were consecutive admissions with a diagnosis of chronic alcohol-induced pancreatitis. They were manual labourers from the periurban areas of Johannesburg. Most were underweight at a mean of 51.9 kg (range, 39.5–63 kg), but all were medically stable and ambulatory at the time they were studied. All were heavy drinkers. Alcohol ingestion was estimated to have averaged 170 g/day (range, 90–300 g/day) for 7–26 years; this intake consisted mainly of commercially available beer but to a lesser extent included home brews and spirituous alcohol. Diabetics had a longer drinking history than nondiabetics: 14.8 versus 10.9 years, respectively. Although diabetics drank more spirits than beer, the converse applied to nondiabetics. According to the patients’ history, all had ceased drinking between approximately 4 weeks and 1 year before commencement of the study.

Diagnostic criteria for the diagnosis of chronic pancreatitis were based on radiographic findings of calcification of the pancreas, and on imaging findings at endoscopic retrograde cholangiopancreatography characteristic of chronic alcoholic pancreatitis [2].

In 12 of the 13 patients, pre-biopsy oral tetracycline double bone labeling was carried out using demethylchlortetracycline 300 mg twice a day for 2 days, and again for another 2 days after a 10-day drug-free interval. Tetracycline double labeling permits measurement of time-based bone formation variables [11]. The bone biopsies were taken between 2 and 7 days after the last dose of tetracycline. Written and informed consent had been obtained from all patients, and the study was approved by the Committee for Research on Human Subjects of the University of the Witwatersrand, Johannesburg, South Africa.

Control transiliac bone samples [12, 13] were obtained from 37 black males aged 37.7 ± 12.3 (median, 35; range, 22–68) years. Their age did not differ significantly from that of the patients (P = 0.76). Eight of the 37 control bone samples were taken from patients who underwent limb surgery for conditions other than a recent fracture or metabolic bone disease (as judged by biochemistry and radiographs), and they had to be ambulant and free of any acute or chronic illness. These 8 subjects had received pre-biopsy double tetracycline labeling according to similar labeling schedules as for the patients. The mean (±SD) age of these tetracycline-labeled controls was 37.3 ± 10.8 years (median, 35; range, 23–54 years) and did not differ from that of the patients (P = 0.5). The remaining 29 control samples were obtained from cadavers of previously healthy individuals who had died suddenly (motor vehicle accidents, train accidents, assaults). We excluded bone samples from subjects who were nonambulant before death, or who at autopsy showed evidence of organic disease (e.g., liver, kidney, gastrointestinal tract, thyroid, parathyroids, lungs) that could have affected bone, or whose death occurred more than 2, or at most 3, days previously.

Bone samples

All patients and control subjects had a transiliac iliac bone biopsy taken from the standard bone biopsy site [14], and bone specimens were processed undecalcified as previously described [12, 13].

Histomorphometric analysis

Cortical (Ct) bone was examined manually by point and intersect counting, linear measurement, and counting of features per unit area [13]; trabecular (Tb) bone was analyzed with the aid of a semiautomatic image analysis system using the Osteoplan program [15]. Nomenclature, abbreviations, and symbols of terms used are those approved by the American Society for Bone and Mineral Research [16]. All specimens were examined by the same investigator (C.M.S.). Intraobserver variability of cortical bone measurements has been reported previously [13]. Iron overload was assessed on unstained sections by counting the number of brown iron granules (hemosiderin-laden macrophages)/mm2 of bone marrow using a calibrated Zeiss Integration Platte II. The presence of more than 6 iron granules/mm2 was considered to represent iron overload as previously defined [17].

Biochemistry

Venous blood specimens were obtained following an overnight fast.

Serum calcium, serum magnesium, and urinary calcium were determined by atomic absorption spectrometry. Serum phosphorus, total alkaline phosphatase, albumin, gamma-glutamyltranspeptidase (GGT), aspartate transaminase (AST), alanine transaminase (ALT), creatinine, and urinary creatinine were measured by Technicon autoanalyser methods; mean corpuscular volume (MCV) of erythrocytes by a Coulter Counter Model S-Plus; parathyroid hormone (PTH) by an immunoradiometric assay for the intact molecule (Nichols Institute, San Juan Capistrano, CA, USA); 25-hydroxyvitamin D (25OHD) and 1,25-dihydroxyvitamin D (1,25(OH)2D) by competitive protein binding assays; leukocyte vitamin C by the method of Gibson and associates [18]; serum iron and total iron-binding capacity (unsaturated iron-binding capacity, UIBC, was calculated as = total iron-binding capacity – serum iron) according to methods previously described [19, 20]; and ferritin by the method of Conradie and Mbhele [21]. Control values were taken from age-matched healthy black males who came to hospital for minor procedures, except for 25OHD, 1,25(OH)2D, and urinary calcium/creatinine values, which were taken from a study of normal black women aged 21–65 years [22].

Statistical analysis

The data were analyzed using the Statistical Analysis System (SAS) program (SAS Institute, Cary, NC, USA). Differences between patients and controls, and between diabetics and nondiabetics, were analyzed by one-way analysis of variance (Tables 1, 2, 3), except for differences between patients and controls in levels of 25OHD, 1,25(OH)2D, and urinary calcium/creatinine ratios, which were tested for the standard error of difference between means. Correlations were tested by the Spearman rank correlation coefficient (Table 4).
Table 1

Histomorphometric variables of iliac crest cortical bone in patients with chronic pancreatitis and in controls

Variables

Patients

Controls

P values

n

Mean ± SD

n

Mean ± SD

Patients versus controls

Median (range)

Median (range)

Structure

 Cortical thickness (Ct.Th, μm)

13

713 ± 396

37

988 ± 330

0.018

643 (284–1558)

940 (455–1593)

 Ec wall thickness (Ec.W.Th, μm)

13

37.3 ± 7

31

44.9 ± 4.9

0.0002

38.1 (26.7–48.7)

45.2 (35.1–56.1)

 Cortical porosity (Ct.Po, %)

13

4.89 ± 3.25

37

4.17 ± 1.49

0.289

4.07 (0.93–13)

3.96 (1.81–7.73)

 Canal number (Ca.N, N/mm2)

13

13.4 ± 5

37

12.5 ± 4.1

0.543

12.9 (6.7–22.9)

12.5 (4.5–22.5)

 Canal diameter (active and inactive) (Ca.Dm, μm)

13

67.5 ± 23.1

37

66.8 ± 15

0.905

65.8 (40.5–108)

65 (40–108)

 Inactive canal diameter (Inactive.Ca.Dm, μm)

13

25.5 ± 4.5

37

27.1 ± 4.5

0.289

25.6 (18.4–32)

26.0 (14.4–38)

 H osteonal diameter (H.On.Dm, μm)

13

124 ± 16

37

129 ± 20

0.395

120 (105–152)

130 (60–169)

 H wall thickness (H.W.Th, μm)

13

49.3 ± 6.7

37

51.1 ± 8.6

0.49

46.6 (41.6–61.2)

50.9 (22.9–67.1)

Static bone turnover

 H osteoid thickness (H.O.Th., μm)

13

7.5 ± 2.2

37

8.9 ± 3.5

0.164

7.1 (3.9–11.5)

8.4 (2.1–16.7)

 H osteoid surface (H.OS/BS, %)

13

9.2 ± 7.6

37

9.5 ± 6.8

0.877

5.3 (2.5–24.4)

7.3 (0.9–28.8)

 H eroded surface (H.ES/BS, %)

13

8.1 ± 7.3

37

5.8 ± 2.9

0.114

6.6 (0–30)

5.4 (1.6–13.1)

 Ec osteoid thickness (Ec.O.Th, μm)

13

7.8 ± 2

37

11.4 ± 3.6

0.0012

8.1 (4.5–11.5)

11.5 (3.9–19.7)

 Ec osteoid surface (Ec.OS/BS, %)

13

12.4 ± 8

37

17.3 ± 10.7

0.138

11.6 (1.6–33.7)

16.8 (3.1–43.9)

 Ec eroded surface (Ec.ES/BS, %)

13

11 ± 8.2

37

6.1 ± 3.2

0.004

10.3 (1.2–26.8)

6.6 (0.4–11.9)

Dynamic bone turnover

 H mineral apposition rate (H.MAR, μm/day)

12

0.68 ± 0.215

8

0.813 ± 0.251

0.222

0.709 (0.364–1.067)

0.765 (0.415–1.186)

 H mineralizing bone surface (H.MS/BS, %)

12

6.8 ± 5.2

8

7.2 ± 3.6

0.866

4 (1.7–15.7)

6.4 (2.8–13.7)

 H bone formation rate (H.BFR/BS, μm3/μm2/year)

12

19.2 ± 18.5

8

22 ± 13.7

0.784

10.7 (3.2–59.6)

19.4 (7–44.5)

 H adjusted apposition rate (H.Aj.AR, μm/day)

12

0.52 ± 0.219

8

0.541 ± 0.206

0.922

0.498 (0.156–1.008)

0.51 (0.26–0.92)

 H mineralization lag time (H.Mlt, days)

12

17.2 ± 8.3

8

21 ± 7

0.295

14.8 (10.4–41)

23.1 (9.8–30.1)

 H activation frequency (H.Ac.f, N/year)

12

0.391 ± 0.354

8

0.425 ± 0.29

0.891

0.214 (0.053–0.975)

0.374 (0.126–0.856)

 Ec mineral apposition rate (Ec.MAR, μm/day)

12

0.66 ± 0.133

8

0.948 ± 0.313

0.011

0.65 (0.53–0.98)

0.873 (0.514–1.375)

 Ec mineralizing bone surface (Ec.MS/BS, %)

12

11 ± 6.4

8

14.8 ± 8.8

0.277

9.1 (2.1–22.3)

14.5 (0.8–26.4)

 Ec bone formation rate (Ec.BFR/BS, μm3/μm2/year)

12

26.7 ± 16.3

8

56.2 ± 40.6

0.035

24.9 (4.3–54)

57.1 (1.5–132.6)

 Ec adjusted apposition rate (Ec.Aj.AR, μm/day)

12

0.58 ± 0.344

8

0.746 ± 0.493

0.386

0.509 (0.141–1.469)

0.728 (0.094–1.571)

 Ec mineralization lag time (Ec.Mlt, days)

12

19 ± 11.7

8

42.4 ± 50.6

0.136

17.9 (4.4–46.4)

18.7 (10.2–145)

 Ec activation frequency (Ec.Ac.f, N/year)

12

0.771 ± 0.559

8

1.21 ± 0.9

0.193

0.647 (0.09–2.025)

1.212 (0.027–3.059)

Statistically significant values are indicated in bold

Ec endocortical, H haversian

Table 2

Histomorphometric variables of iliac crest trabecular (Tb) bone in patients with chronic pancreatitis and in controls

Variables

Patients

Controls

P values

n

Mean ± SD

n

Mean ± SD

Patients versus controls

Median (range)

Median (range)

Structure

 Tb bone volume (BV/TV, %)

13

14.9 ± 8.1

37

20.1 ± 6.1

0.019

13.3 (5.8–35.3)

18.6 (11.4–36.4)

 Tb thickness (Tb.Th, μm)

13

108 ± 61

37

161 ± 43

0.001

94 (58–302)

165 (86–258)

 Tb wall thickness (Tb.W.Th, μm)

13

31.7 ± 5.3

35

46.1 ± 4.9

<0.0001

32 (20.1–41.9)

45.9 (39.2–59.2)

 Tb number (Tb.N, N/mm)

13

1.41 ± 0.48

37

1.29 ± 0.34

0.332

1.39 (0.61–2.12)

1.3 (0.62–1.89)

 Tb separation (Tb.Sp, μm)

13

701 ± 339

37

676 ± 238

0.773

568 (385–1540)

607 (374–1423)

 Iron granules in marrow (Irgr, N/mm2)

13

32 ± 46.2

37

0.69 ± 1.69

0.0001

8.4 (0–140)

0 (0–6)

Static bone turnover

 Tb osteoid thickness (Tb.O.Th, μm)

13

7.4 ± 2.8

37

8.81 ± 2.9

0.14

8.5 (3.1–11.9)

8.3 (2.7–14.7)

 Tb osteoid surface (Tb.OS/BS, %)

13

13.8 ± 6.8

37

17.6 ± 12.3

0.295

14.2 (1.9–24)

14.1 (3.9–46.9)

 Tb osteoid volume (Tb.OV/BV, %)

13

2.23 ± 1.37

37

2.47 ± 1.9

0.671

2.21 (0.13–4.76)

1.93 (0.33–7.14)

 Tb eroded surface (Tb.ES/BS, %)

13

5.7 ± 4.2

37

4.7 ± 2.3

0.306

3.6 (1.8–16.4)

4.1 (1.3–11.0)

 Tb mean erosion depth (Tb.E.De, μm)

13

6 ± 2.4

33

7.1 ± 2

0.097

5 (4–11.4)

7.1 (4–13)

Dynamic bone turnover

 Tb mineral apposition rate (Tb.MAR, μm/day)

12

0.622 ± 0.11

8

0.736 ± 0.264

0.195

0.637 (0.425–0.777)

0.745 (0.345–0.152)

 Tb mineralizing surface (Tb.MS/BS, %)

12

6.6 ± 4.2

8

5.1 ± 3.5

0.412

6.3 (0.7–13.8)

4.1 (0.7–11.1)

 Tb bone formation rate (Tb.BFR/BS, μm3/μm2/year)

12

15.8 ± 11.7

8

13.9 ± 10

0.719

13.8 (1.5–39)

12.9 (0.9–33.4)

 Tb adjusted apposition rate (Tb.Aj.AR, μm/day)

12

0.278 ± 0.118

8

0.322 ± 0.181

0.52

0.297 (0.05–0.475)

0.312 (0.023–0.676)

 Tb mineralization lag time (Tb.Mlt, days)

12

34 ± 25

8

39 ± 33

0.72

28 (14–108)

30 (9.8–115.4)

 Tb activation frequency (Tb.Ac.f, N/year)

12

0.466 ± 0.29

8

0.312 ± 0.212

0.213

0.417 (0.05–0.929)

0.3 (0.021–0.719)

Statistically significant values are indicated in bold

Table 3

Biochemical variables in patients with chronic pancreatitis compared to controls

Variables

Patients

Controls

P values

n

Mean ± SD

n

95% CI (mean ± SD)

Median (range)

Bone related

 Calcium (mmol/l)

13

2.5 ± 0.14

30

2.39–2.47 (2.43 ± 0.1)

0.07

0.47 (2.34–2.81)

 Phosphorus (mmol/l)

13

1.34 ± 0.24

30

1.16–1.32 (1.24–0.2)

0.186

1.39 (0.8–1.65)

 Magnesium (mmol/l)

13

0.821 ± 0.096

12

0.862–0.948 (0.91 ± 0.07)

0.02

0.84 (0.68–0.94)

 PTH (pg/ml)

11

13.9 ± 9.2

12

10.1–19.6 (14.84 ± 7.6)

0.793

20.3 (0.7–24.2)

 25OHD (ng/ml)

13

13.5 ± 7.3

107

19–22 (20.8 ± 6.7)

<0.005

13.4 (4–28.8)

 1,25(OH)2 D (pg/ml)

13

33.6 ± 18.5

107

32–36 (34.2 ± 10.1)

NS

32.5 (9.4–77.1)

 Ca/creatinine (mmol/mmol)

13

0.278 ± 0.217

149

0.146–0.176 (0.161 ± 0.095)

1 > P > 0.05

0.241 (0.021–0.708)

Alcohol- and liver related

 MCV (fl)

13

97 ± 4

17

85–90 (88 ± 5)

<0.0001

98 (90–101)

 GGT (U/l)

13

362 ± 518 112 (33–1920)

18

30–50 (40 ± 19)

0.013

 AST (U/l)

13

62 ± 56

16

19–38 (29 ± 18)

0.031

46 (14–231)

 Total alkaline phosphatase (U/l)

13

287 ± 489

30

90–122 (106 ± 42)

0.048

120 (50–1870)

 ALT (U/l)

13

38.6 ± 41.6

18

21–40 (31 ± 19)

0.484

29 (5–170)

 Albumin (g/l)

13

40.4 ± 4.8

18

40–44 (42 ± 4.5)

0.347

39 (32–51)

Iron related

 Iron (μg/100 ml)

12

93 ± 28

18

53–87 (70 ± 34)

0.059

86 (54–148)

 UIBC (μg/100 ml)

12

161 ± 40

18

175–215 (195 ± 40)

0.029

163.5 (78–233)

 Ferritin (μg/l)

12

640 ± 693

18

157–294 (226 ± 138)

0.019

371 (26–2371)

 Ascorbic acid μg/108 WBC

13

15.7 ± 6

17

17–25 (21.2 ± 8)

0.049

15.2 (7–28)

Statistically significant values are indicated in bold

PTH parathyroid hormone, 25OHD 25-hydroxyvitamin D, 1,25(OH)2D 1,25-dihydroxyvitamin D, MCV mean corpuscular volume of erythrocytes, GGT gamma-glutamyl transpeptidase, AST aspartase transaminases, UIBC unsaturated iron-binding capacity, ALT alanine transaminase

Table 4

Correlations between biochemical (alcohol/liver- and iron-related) variables versus bone histomorphometric data in 13 patients with chronic pancreatitis

Variable versus variable

r

P values

Alcohol/liver related

 MCV

  Tb osteoid thickness

0.593

0.033

 GGT

  Ec double-labeled surface

−0.601

0.039

 AST

  Tb thickness

0.68

0.011

  Ec mineral apposition rate

0.664

0.019

 ALT

  Tb thickness

0.84

0.0003

  Ec mineral apposition rate

0.677

0.016

 Alkaline phosphatase

  Cortical thickness

−0.619

0.024

  H wall thickness

−0.58

0.038

  Tb wall thickness

−0.564

0.045

  Ec mineralizing surface

−0.699

0.011

 Albumin

  Ec mineralizing surface

0.756

0.005

  Ec bone formation rate

0.837

0.0007

  Ec activation frequency

0.833

0.0008

  Tb eroded surface

−0.646

0.017

Iron related

 Iron

  Tb bone formation rate

0.655

0.029

 UIBC

  Ec osteoid surface

−0.664

0.019

  Ec osteoid thickness

−0.734

0.007

  Tb osteoid thickness

−0.797

0.002

  Tb bone formation rate

−0.627

0.039

  Tb activation frequency

−0.645

0.032

 Iron granules

  Tb thickness

0.584

0.036

Only correlations with significant P values are given. Statistically significant values are indicated in bold

Results

Histomorphometry

Results of the histomorphometric analysis of cortical (Ct) and trabecular (Tb) bone are given in Tables 1 and 2 and Figs. 1 and 2.
Fig. 1

Scatterplot of cortical structural and bone turnover variables that differed between patients with chronic pancreatitis and controls

Fig. 2

Scatterplot of trabecular structural and bone turnover variables that differed between patients with chronic pancreatitis and controls

Bone structure

Ct bone (Table 1, Fig. 1) in the patients showed lower values for Ct thickness and Ec wall thickness than in controls. Tb bone (Table 2, Fig. 2) showed lower values for bone volume, Tb thickness, and Tb wall thickness. Increased numbers of hemosiderin-laden macrophages were present in the bone marrow of 7 (all nondiabetics) of the 13 patients, the mean number of hemosiderin-laden macrophages in the bone marrow being greater in patients than in controls.

Bone turnover

Patients had thinner Ec osteoid seams, lower Ec mineral apposition rate, lower Ec bone formation rate, and more extensive Ec eroded surface than controls (Table 1; Fig. 1). Tb bone turnover data did not differ significantly, but Tb osteoid thickness and mineral apposition rate tended to be lower in patients compared to controls (Table 2, Fig. 2).

Diabetics versus nondiabetics

Diabetics (n = 6) compared to nondiabetics (n = 7) showed no statistically significant differences. However, diabetics showed a trend to lower values for Ec wall thickness (34.4 ± 6.5 vs. 39.8 ± 6.8 μm; P = 0.169), Tb bone volume (11% ± 4.9% vs. 18.3% ± 9.1%; P = 0.107), Tb osteoid thickness (6.2 ± 2.8 vs. 8.5 ± 2.5 μm; P = 0.144), H eroded surface (4.8% ± 3.1% vs. 10.8% ± 8.8%; P = 0.146), Tb erosion depth (4.8 ± 0.8 vs. 7 ± 3 μm; P = 0.104), and hemosiderin-laden macrophages in bone marrow (7 ± 12 vs. 53 ± 55 N/mm2; P = 0.068). All other variables had even less significant P values.

Biochemistry

Results of biochemical variables are given in Table 3. All patients had normal serum creatinine levels.

Bone-related variables

Serum 25OHD and serum magnesium levels were lower in patients than in controls. However, patients had normal mean concentrations of 1,25(OH)2D and PTH. Urinary calcium/creatinine values tended to be higher in patients than in controls.

Alcohol/liver-related variables

Markers of alcohol abuse—MCV, GGT, and AST—[23] were elevated in patients compared to controls. One patient, a nondiabetic, had normal liver function tests. All other patients had between one and five abnormal values (GGT, AST, ALT, alkaline phosphatase, albumin). Mean values of total alkaline phosphatase were also elevated, but albumin and ALT were normal.

Iron-related variables

Seven of the 13 patients showed evidence of iron overload in the form of increased numbers of hemosiderin-laden macrophages in the bone marrow. Mean levels of serum iron and ferritin were elevated, and UIBC and ascorbic acid levels were low in the patients compared to controls.

Diabetics versus nondiabetics

Five of the 6 diabetic patients and 1 of the nondiabetic patients had reduced concentrations of 1,25(OH)2D compared to controls: mean values were diabetics 22 ± 7.9 and nondiabetics 43.6 ± 19.5 pg/ml (P = 0.028). The 6 diabetic (100%) and 5 of the 7 (71%) nondiabetic patients had subnormal 25(OH)D levels (diabetics 10 ± 4.5 vs. nondiabetics 16.5 ± 8.2 ng/ml), but this difference was not statistically significant (P = 0.155). Hypomagnesemia was found in 5 of the 6 (83.3%) diabetics and in 4 of the 7 (57.1%) nondiabetics (diabetics 0.792 ± 0.115 vs. nondiabetics 0.846 ± 0.076 mmol/l, respectively; P = 0.333). Neither did diabetics differ from nondiabetics with respect to other biochemical variables.

Correlations

MCV, AST, and ALT showed positive correlations and GGT and alkaline phosphatase negative correlations with some structural and bone formation variables (Table 4). Albumin was positively correlated with several Ec formation variables and negatively with Tb eroded surface. Iron-related variables were positively correlated and UIBC negatively correlated with variables of bone structure and formation. Urinary calcium/creatinine ratio was negatively correlated with Tb bone formation rate (r = –0.685, P = 0.014), and urinary magnesium/creatinine ratio was positively correlated with Tb eroded surface (r = 0.735, P = 0.004) and Tb erosion depth (r = 0.626, P = 0.022).

Discussion

This study of cortical and trabecular bone in patients with chronic pancreatitis caused by alcohol abuse showed loss of cortical thickness and trabecular bone volume together with microarchitectural deterioration in both bone compartments compared to controls. The thinner Ec osteoid seams and lower Ec mineral apposition rate and lower Ec bone formation rate reflect impaired bone formation in the patients compared to controls, and these changes may be expected to have led to lower Ec wall thickness and through that to reduced cortical thickness. Although lower mineral apposition rate reflects diminished vigor of osteoblasts at the individual cellular level [24], lower bone formation rate is an indicator of lower osteoblast team performance [25]. Reduction of this latter calculated variable (bone formation rate = mineral apposition rate × mineralizing surface) is thus contributed to by both diminished osteoblast vigor and—because mineralizing surface depends on osteoblast numbers—by impaired osteoblast recruitment [25]. In the presence of impaired Ec bone formation, increased Ec eroded surface suggests uncoupling of formation from resorption in favor of resorption. Such uncoupling is expected to lead to imbalance at the remodeling cycle in that less new bone is formed than was resorbed. This imbalance is expected to lead to accelerated Ec bone loss and cortical thinning.

The decrease in Tb bone volume was characterized by a reduction in Tb thickness but not in Tb number. Consequently, there was no increase in Tb separation. Lower Tb wall thickness, and ultimately reduced Tb thickness, may have resulted from a tendency to lower Tb mineral apposition rate compared to controls. As our patients showed no increase in Tb eroded surface or in Tb erosion depth, it may be assumed that Tb thinning came about mainly through deficient bone formation. Tb bone in the present study therefore exhibits a state of low bone turnover. This form of bone loss thus differs from high turnover osteoporosis in which increased eroded surface and increased erosion depth lead to Tb plate perforations [26, 27], so that Tb numbers decline and Tb separation increases. By contrast, in the case of low trabecular bone turnover with unaltered Tb eroded surface and Tb erosion depth, as in the present study, plate perforations are less likely to occur; instead, trabeculae became thinner but trabecular numbers and trabecular separation were little changed.

To sum up, on both the Ec and the Tb envelopes bone loss was brought about by remodeling imbalance: on the Ec envelope through impaired formation and enhanced resorption, and on the Tb envelope mainly through impaired formation in the presence of unchanged resorption. An important finding is that none of the patients showed evidence of osteomalacia by accepted criteria [28].

Bone disease associated with alcohol-induced pancreatitis is likely the result of multifactorial pathogenesis. In addition to the adverse effects of alcohol abuse and chronic pancreatitis, malabsorption, vitamin D deficiency, diabetes mellitus, liver disease, poor nutrition, vitamin C deficiency, and possibly iron overload may play a role.

The effects of alcohol on bone have been well documented in the literature, although chronic pancreatitis is rarely mentioned in this context. There is consensus that excessive intake of alcohol leads to bone loss through reduced bone formation [5, 6, 8, 9, 29, 30, 31, 32], whereas eroded bone surface may [5, 9, 17] or may not be elevated [6]. Tb erosion depth was found not to be increased in the present and other studies [9, 17]; if anything, Tb erosion depth tended to be lower in the present study compared to controls. The depressant effect of alcohol ingestion on osteoblast function has been found to be dose dependent [6] and immediate [33]. Tb thinning [9, 29, 34] and reduced Tb wall thickness [9, 29], as found in the present study, have also been reported in other investigations of alcoholics with osteoporosis. In rats and mice fed alcohol, trabecular thinning was also demonstrated and found to be dose dependent [6, 35]. The trabecular histomorphometric features of the bone disease in the present study are thus consistent with those of alcohol-induced bone disease as described in the literature. In addition, this study describes the mode of bone loss in cortical bone in this condition.

Significant bone disease resulting from isolated pancreatic insufficiency is thought to be uncommon [36], and studies on the effects on bone in chronic pancreatitis not induced by alcohol abuse are scarce. One such study by Mann and associates [1] reported low levels of 25(OH)D and 1,25(OH)2D and diminished bone mineral density (BMD). PTH and biochemical markers of bone turnover were normal in that study. The authors were able to demonstrate that a decline in serum vitamin D metabolite concentrations and BMD worsened with increasing severity of pancreatic morphological abnormalities and diminishing exocrine function. They attributed bone loss to a number of factors, namely, vitamin D deficiency (resulting from malabsorption), physical inactivity, and a possible systemic effect of chronic inflammation on bone. Similar findings of low BMD and decreased vitamin D metabolites in chronic pancreatitis have also been described by others [4, 37], but their studies included patients abusing alcohol. Bone loss reported in nonalcoholic chronic pancreatitis did not exceed 10% relative to controls [1]; however, in our patients bone loss was more severe in that cortical thickness was reduced by 28% and trabecular bone volume by 26% compared to controls.

Vitamin D deficiency is known to lead to bone loss [38]. Even low normal concentrations of vitamin D metabolites have been found to adversely affect bone mass [39]. The low levels of 25(OH)D in the present study, which may have resulted from low intake of vitamin D, malabsorption, or impaired hepatic hydroxylation could have contributed to bone loss; however, there is no evidence in the present data for such an association.

Insulin-dependent diabetes is a known risk factor for osteopenia [40]. The diabetic patients tended to have greater deterioration of bone structure and turnover than the nondiabetic patients. Although these differences were not statistically significant (small numbers), they point to diabetes mellitus being an aggravating factor. The lower levels of 25(OH)D and 1,25(OH)2D in the diabetic patients (significant only for 1,25(OH)2D) may have contributed further to bone loss in the diabetics. Moreover, the longer drinking history and the preference for spirits of the diabetics may have adversely affected both the pancreas and bone. The preference for beer in nondiabetics is reflected in a greater number of hemosiderin-laden macrophages in bone marrow, attributable to consumption of traditional beer with high iron content [17].

Alcoholic liver disease appeared to have had a limited adverse effect on bone in view of negative correlations between only GGT and alkaline phosphatase versus bone structural and formation variables, although abstinence from alcohol intake before the investigation could have improved some liver function tests. It is more likely that alcohol had a direct toxic effect on bone. It is noteworthy that with the exception of one low value serum albumin levels in this alcoholic cohort were normal, which, apart from being an indicator of hepatic synthetic capacity, also indicates the absence of any significant degree of malnutrition despite low body weight.

Iron overload resulting from consumption of traditional beer brewed in iron vessels [41] was mild to moderate in the six patients affected and did not appear to have had adverse effects on bone, because significant correlations between indicators of iron overload versus bone structure and bone formation were positive and those with UIBC negative (see Table 4). This result is in keeping with our previous findings that iron in its storage forms, ferritin and hemosiderin, is not itself directly toxic to bone but rather that bone loss in iron overload is related to ascorbic acid deficiency as a result of iron-mediated catabolism of ascorbic acid [17, 42]. Low ascorbic acid levels as found in the present study have been previously documented in black South Africans with chronic alcoholic pancreatitis and were attributed to low dietary intake [43]. The low ascorbic acid concentrations in the present study were probably not caused by the mild iron overload because ascorbic acid levels did not significantly correlate with variables reflecting iron overload. Nor did ascorbic acid deficiency appear to play a major role in the bone disease, as there were no significant correlations between ascorbic acid levels and histomorphometric variables. Elevation of levels of serum ferritin may have been enhanced by alcohol consumption.

The elevated levels of urinary calcium excretion and their negative correlation with Tb bone formation may also be an effect of alcohol abuse [44]. Lower serum magnesium levels in the patients could have been caused by any of a number of factors such as alcohol abuse, diabetes, malabsorption, steatorrhea, and diarrhea [45]; it tended to be more common in diabetic (83%) than nondiabetic patients (57%). The mild degree of hypomagnesemia, however, did not result in a reduction in either PTH or serum calcium levels.

In the pathogenesis of the bone disease of alcohol-induced pancreatitis, alcohol itself appeared to have been the most detrimental factor, with secondary adverse effects of liver disease, reduced vitamin D metabolites, diabetes mellitus, and possibly pancreatitis and malabsorption. Alcohol toxicity also best describes the bone disease in the present study. Apart from its high level of toxicity to bone, alcohol was after all the first offender in time in the long line of pathogenetic factors responsible for this bone disease.

Limitations

Limitations of this study were small patient numbers, lack of staging of morphological pancreatic abnormalities, and use of cadaver bone samples among the controls.

Despite the small number of patients, this may be the most detailed histomorphometric study of cortical and trabecular bone in patients with chronic alcoholic pancreatitis to date. The findings are consistent with alcohol bone disease. Osteomalacia was not a feature. Other second-line pathogenetic factors were liver disease, hypovitaminosis D, diabetes mellitus, and possibly chronic pancreatitis.

Notes

Acknowledgments

This study was funded by the Medical Research Council of South Africa, and the University of the Witwatersrand, Johannesburg, South Africa.

Conflict of interest statement

None.

References

  1. 1.
    Mann STW, Stracke H, Lange U, Klör HU, Teichmann J (2003) Alterations of bone mineral density and bone metabolism in patients with various grades of chronic pancreatitis. Metabolism 52:579–585CrossRefPubMedGoogle Scholar
  2. 2.
    Segal I, Lerios M, MacPhail AP, Di Bisceglie AM, Grieve T (1988) Genesis of chronic pancreatitis in the South African black population. S Afr Med J 74:385–386PubMedGoogle Scholar
  3. 3.
    Nair RJ, Lawler L, Miller WR (2007) Chronic pancreatitis. Am Fam Physician 76:1679–1688PubMedGoogle Scholar
  4. 4.
    Haaber AB, Rosenfalck AM, Hansen B, Hilsted J, Larsen S (2000) Bone mineral metabolism, bone mineral density, and body composition in patients with chronic pancreatitis and pancreatic exocrine insufficiency. Int J Pancreatol 27:21–27CrossRefPubMedGoogle Scholar
  5. 5.
    Arlot ME, Bonjean M, Chavassieux PM, Meunier PJ (1983) Bone histology in adults with aseptic necrosis. J Bone Joint Surg (Am) 65:1319–1327Google Scholar
  6. 6.
    Turner RT, Kidder LS, Kennedy A, Evans GL, Sibonga JD (2001) Moderate alcohol consumption suppresses bone turnover in adult female rats. J Bone Miner Res 16:589–594CrossRefPubMedGoogle Scholar
  7. 7.
    Seeman E (1996) Nutrition and risk for osteoporosis. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis. Academic Press, San Diego, pp 577–597Google Scholar
  8. 8.
    Crilly RG, Anderson C, Hogan D, Delaquerrière-Richardson L (1988) Bone histomorphometry, bone mass, and related parameters in alcoholic males. Calcif Tissue Int 43:269–276CrossRefPubMedGoogle Scholar
  9. 9.
    Chappard D, Plantard B, Petitjean M, Alexandre C, Riffat G (1991) Alcoholic cirrhosis and osteoporosis in men: a light and scanning microscopy study. J Stud Alcohol 52:269–274PubMedGoogle Scholar
  10. 10.
    Owor R (1972) Quantitative estimation of bone mass in Africans with particular reference to bone changes in chronic pancreatic disease. East Afr Med J 49:860–867PubMedGoogle Scholar
  11. 11.
    Frost HM (1983) Bone histomorphometry: choice of marking agent and labeling schedule. In: Recker RR (ed) Bone histomorphometry: techniques and interpretation. CRC Press, Boca Raton, pp 37–52Google Scholar
  12. 12.
    Schnitzler CM, Pettifor JM, Mesquita JM, Bird MDT, Schnaid E, Smyth AE (1990) Histomorphometry of iliac crest bone in 346 normal black and white South African adults. Bone Miner 10:183–199CrossRefPubMedGoogle Scholar
  13. 13.
    Schnitzler CM, Mesquita JM (2006) Cortical bone histomorphometry of the iliac crest in normal black and white South African adults. Calcif Tissue Int 79:373–382CrossRefPubMedGoogle Scholar
  14. 14.
    Melsen F, Mosekilde L (1981) The role of bone biopsy in the diagnosis of metabolic bone disease. Orthop Clin N Am 12:571–602Google Scholar
  15. 15.
    Malluche HH, Meyer W, Sherman D, Massry SC (1982) Quantitative bone histology in 84 normal American subjects. Calcif Tissue Int 34:449–455CrossRefPubMedGoogle Scholar
  16. 16.
    Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR (1987) Bone histomorphometry: standardization of nomenclature, symbols and units. J Bone Miner Res 2:595–610CrossRefPubMedGoogle Scholar
  17. 17.
    Schnitzler CM, MacPhail AP, Shires R, Schnaid E, Mesquita JM, Robson HJ (1994) Osteoporosis in African hemosiderosis: role of alcohol and iron. J Bone Miner Res 9:1865–1873CrossRefPubMedGoogle Scholar
  18. 18.
    Gibson SLM, Moore FML, Goldberg A (1966) Measurement of leucocyte ascorbic acid. BMJ 1:1152–1153CrossRefPubMedGoogle Scholar
  19. 19.
    International Committee for Standardization in Haematology (1978) Recommendations for measurement of serum iron in human blood. Br J Haematol 38:291–294CrossRefGoogle Scholar
  20. 20.
    International Committee for Standardization in Haematology (1978) The measurement of total and unsaturated iron-binding capacity in serum. Br J Haematol 38:281–287CrossRefGoogle Scholar
  21. 21.
    Conradie JD, Mbhele BE (1980) Quantitation of serum ferritin by enzyme linked immunosorbent assay (ELISA). S Afr Med J 57:282–287PubMedGoogle Scholar
  22. 22.
    Daniels ED, Pettifor JM, Schnitzler CM, Moodley GP, Zachen D (1979) Differences in mineral homeostasis, volumetric bone mass and femoral neck axis length in black and white South African women. Osteoporosis Int 7:105–112CrossRefGoogle Scholar
  23. 23.
    Schnitzler CM, Menashe L, Sutton CG, Sweet MBE (1988) Serum biochemical and haematological markers of alcohol abuse in patients with femoral neck and intertrochanteric fractures. Alcohol Alcohol 23:127–132PubMedGoogle Scholar
  24. 24.
    Parfitt AM (1994) Osteonal and hemi-osteonal remodeling: the spatial and temporal framework for signal traffic in adult human bone. J Cell Biochem 55:273–286CrossRefPubMedGoogle Scholar
  25. 25.
    Parfitt AM, Villanueva AR, Foldes J, Sudhaker Rao D (1995) Relations between histologic indices of bone formation: implications for the pathogenesis of spinal osteoporosis. J Bone Miner Res 10:466–473CrossRefPubMedGoogle Scholar
  26. 26.
    Parfitt AM, Mathews CHE, Villanueva AR, Kleerekoper M, Frame B, Rao DS (1983) Relationships between surface, volume, and thickness of iliac trabecular bone in aging and in osteoporosis. J Clin Invest 72:1396–1409CrossRefPubMedGoogle Scholar
  27. 27.
    Dempster DW (1995) Bone remodeling. In: Riggs BL, Melton LJ III (eds) Osteoporosis. Lippincott-Raven, Philadelphia, pp 67–91Google Scholar
  28. 28.
    Parfitt AM (1998) Osteomalacia and related disorders. In: Avioli LV, Krane SM (eds) Metabolic bone disease. Academic Press, San Diego, pp 327–386Google Scholar
  29. 29.
    De Vernejoul MC, Bielakoff HerveM, Gueris J, Hott M, Modrowski D, Kuntz D, Miravet L, Ryckewaert A (1983) Evidence for defective osteoblast function: a role for alcohol and tobacco consumption in middle-aged men. Clin Orthop Relat Res 179:107–115PubMedGoogle Scholar
  30. 30.
    Shankar K, Hidestrand M, Liu X, Chen JR, Haley R, Perrien DS, Skinner RA, Lumpkin CK Jr, Badger TM, Ronis MJJ (2008) Chronic alcohol consumption inhibits postlactational anabolic bone rebuilding in female rats. J Bone Miner Res 23:338–349CrossRefPubMedGoogle Scholar
  31. 31.
    Iwaniec UT, Trevisiol CH, Maddalozzo GF, Rosen CJ, Turner RT (2008) Effects of low dose parathyroid hormone on bone mass, turnover, and ectopic osteoinduction in a rat model for chronic alcohol abuse. Bone (NY) 42:695–701Google Scholar
  32. 32.
    Bikle DD, Genant HK, Cann C, Recker RR, Halloran BP, Strewler GJ (1985) Bone diseases in alcohol abuse. Ann Intern Med 103:42–48PubMedGoogle Scholar
  33. 33.
    Nielsen HK, Lundby L, Rasmussen K, Charles P, Hansen C (1990) Alcohol decreases serum osteocalcin in a dose-dependent way in normal subjects. Calcif Tissue Int 46:173–178CrossRefPubMedGoogle Scholar
  34. 34.
    Schnitzler CM, Solomon L (1984) Bone changes after alcohol abuse. S Afr Med J 66:730–734PubMedGoogle Scholar
  35. 35.
    Zhang J, Dai J, Lin D, Habib P, Smith P, Murtha J, Fu Z, Yao Z, Qi T, Keller ET (2002) Osteoprotegerin abrogates chronic alcohol ingestion-induced bone loss in mice. J Bone Miner Res 17:1256–1263CrossRefPubMedGoogle Scholar
  36. 36.
    Bikle DD (2007) Vitamin D insufficiency/deficiency in gastrointestinal disorders. J Bone Miner Res 22:V50–V54CrossRefPubMedGoogle Scholar
  37. 37.
    Moran CE, Sosa EG, Martinez SM, Geldern P, Messina D, Russo A, Boerr L, Bal JC (1997) Bone mineral density in patients with pancreatic insufficiency and steatorrhea. Am J Gastroenterol 92:867–871PubMedGoogle Scholar
  38. 38.
    Rosen CJ, Morrison A, Zhou H, Storm D, Hunter SJ, Musgrave K, Chen T, Wen-Wei HolickMF (1994) Elderly women in Northern England exhibit seasonal changes in bone mineral density and calcitropic hormones. Bone Miner 25:83–92CrossRefPubMedGoogle Scholar
  39. 39.
    Chapuy MC, Chapuy P, Thomas JL, Hazard MC, Meunier PJ (1996) Biochemical effects of calcium and vitamin D supplementation in elderly, institutionalized, vitamin D deficient patients. Rev Rhum Engl Ed 63:135–140PubMedGoogle Scholar
  40. 40.
    Shane E (1996) Osteoporosis associated with illnesses and medications. In: Marcus R, Feldman D, Kelsey J (eds) Osteoporosis. Academic Press, San Diego, pp 925–946Google Scholar
  41. 41.
    Seftel HC, Malkin C, Schmaman A, Abrahams C, Lynch SR, Charlton RW, Bothwell TH (1966) Osteoporosis, scurvy, and siderosis on Johannesburg Bantu. BMJ 1:642–646CrossRefPubMedGoogle Scholar
  42. 42.
    Schnitzler CM, Schnaid E, MacPhail AP, Mesquita JM, Robson HJ (2005) Ascorbic acid deficiency, iron overload and alcohol abuse underlie the severe osteoporosis in black African patients with hip fractures—a bone histomorphometric study. Calcif Tissue Int 76:79–89CrossRefPubMedGoogle Scholar
  43. 43.
    Segal I (1998) Pancreatitis in Soweto, South Africa: focus on alcohol-related disease. Digestion 59:25–35CrossRefPubMedGoogle Scholar
  44. 44.
    Kalbfleisch JM, Lindeman RD, Ginn HE, Smith WO (1963) Effects of ethanol administration on urinary excretion of magnesium and other electrolytes in alcoholic and normal subjects. J Clin Invest 42:1471–1475CrossRefPubMedGoogle Scholar
  45. 45.
    Rude RK (1998) Magnesium deficiency: a cause of heterogeneous disease in humans. J Bone Miner Res 13:749–758CrossRefPubMedGoogle Scholar

Copyright information

© The Japanese Society for Bone and Mineral Research and Springer 2010

Authors and Affiliations

  • Christine M. Schnitzler
    • 1
    • 4
  • Julia M. Mesquita
    • 1
    • 2
  • Roy Shires
    • 3
  1. 1.MRC Mineral Metabolism Research UnitUniversity of the WitwatersrandJohannesburgSouth Africa
  2. 2.Division of Orthopaedic SurgeryUniversity of the WitwatersrandJohannesburgSouth Africa
  3. 3.Department of Internal MedicineChris Hani-Baragwanath Hospital, University of the WitwatersrandJohannesburgSouth Africa
  4. 4.Somerset WestSouth Africa

Personalised recommendations