Abnormal Mineral Metabolism in Diabetic Patients -Relevance to Micro and Macrovascular Disease and to Bone Metabolic Disorder
João Martin Martins1,3*, Patrícia Aranha2
1Department of Endocrine, Hospital Santa Maria, Lisbon, Portugal
2Department of Internal Medicine, Hospital Santa Maria, Lisbon, Portugal
3Lisbon Medical School, Lisbon, Portugal
*Corresponding
author: João Martin Martins, Department of Endocrinology, Floor 6, Hospital of Santa Maria by Prof Egas Moniz, 1649-028 Lisboa, Portugal. Tel: +351914402121; Email: jmartinmartins@sapo.pt
Citation: Martins JM, Aranha P
(2018) Abnormal Mineral Metabolism in Diabetic Patients -Relevance to Micro and
Macrovascular Disease and to Bone Metabolic Disorder. J Diabetes Treat: 142. DOI: 10.29011/2574-7568.000042
Background: Diabetic patients present an increased fracture risk despite an
increased bone mineral density; serum Parathyroid Hormone (PTH), phosphate and
25-hydroxivitamin D (25OHD) are now considered independent cardiovascular risk
factors. We systematically evaluated mineral metabolism to assess these
contrasting and emerging factors.
Patients and Methods: We studied 350 diabetic patients. Besides standard clinical care,
serum calcium, phosphate, magnesium, PTH and 25OHD were measured. The
Statistical Package for the Social Sciences Program (IBM SPSS 24th version) was used to
analyze the data.
Results: Mineral metabolism was commonly abnormal (13%). Insulin and
thiazide diuretics alter serum calcium levels. Most patients (85%) presented
low 25OHD levels, inversely related to metabolic control and to PTH levels and
21% presented increased PTH levels. Serum PTH levels were higher in patients
with retinopathy and in those with high blood pressure.
Discussion: Because of nephropathy, diuretic and insulin use, diabetic
patients commonly present abnormal mineral metabolism. Almost all diabetic
patients are deficient/insufficient regarding 25OHD levels in relation to
metabolic control. Secondary hyperparathyroidism and low levels of 25OHD, a
unique combination in diabetic patients may contribute to the increased bone
fragility in diabetic patients and may be a risk factor for micro- and
macrovascular disease.
Keywords: Brittle Bones; Low Vitamin D; Micro and Macrovascular Disease;
Mineral Metabolism; Secondary Hyperparathyroidism
Introduction
Diabetes Mellitus (DM) is one of the most common chronic diseases
in the western world, with a prevalence between 8-10% in the general
population, that is rapidly increasing [1].
Because of long term biochemical abnormalities and micro- and macrovascular
disease, DM is a systemic disease with the widespread organ involvement [2,3]. Because of long term
use of multiple medications - antidiabetic, anti-hypertensive, hypolipidemic,
and anti-platelet drugs - secondary adverse effects are common [4].
An increased risk of bone fractures has been reported in diabetic
patients - up to six-fold in Type 1 Diabetics (DM1) and two-fold in Type 2
Diabetics (DM2) - that does not seem to be accounted for by an increased risk
of falls and everyday trauma due to hypoglycemia, reduced visual acuity and
peripheral or autonomic neuropathy (5-6). Despite this, Bone Mineral Density
(BMD) although reduced in DM1 has been consistently reported as increased in
the much larger group of DM2 [7-9]. This apparent paradox suggests decreased bone quality in DM
patients although the mechanisms remain largely speculative [10-14].
Minerals accumulate in bone (99%, 85% and around 50% of total body
content for calcium, phosphate and magnesium respectively) and together they
comprise more than 65% of the skeleton weight (15). Data regarding mineral
metabolism is therefore relevant to understand bone quality, bone mineral
density and fracture risk in diabetic patients.
Abnormal mineral metabolism is common in the general population
with an estimated prevalence of 1-3% for primary hyperparathyroidism, more than
50% for vitamin D insufficiency/deficiency and 9% for nephrolithiasis [16]. Hyper or hypocalcemia
are also relatively common in the general population and may occur in 1% and
25% of hospital admissions respectively [17]. Disordered mineral metabolism may be even more common in DM
patients; because of autonomic neuropathy, mineral intake, gastrointestinal
absorption and excretion may be altered, while because of nephropathy mineral
losses may be increased, and because of obesity and mobility limitations
mineral distribution may be different [2,3,18-20]. Also, inhibitors of Sodium Glucose Linked Transporters (SGLT-2),
insulin, thiazide and loop diuretics may alter the renal handling of water,
sodium and minerals [4,21].
More recently, serum PTH is emerging as an independent risk factor
for cardiovascular disease and all-cause mortality, although the relation is
stronger with the former [22,23]. Since the increased risk is apparent in either primary or
secondary hyperparathyroidism, Chronic Kidney Disease (CKD) cannot be the only
explanation. In fact, the increased risk occurs within the normal range of
serum PTH and persists after correction for the glomerular filtration rate.
This effect is a major one with a one standard deviation increase in serum PTH
being associated with a 30-40% increase in cardiovascular risk and in fact is
equivalent to other well-established risk factors for cardiovascular disease
like systolic blood pressure and total cholesterol. These data are further
reinforced by intervention studies with calcimimetic drugs that reduce the
cardiovascular risk [22]. A direct effect of PTH on vascular stiffening, remodeling,
calcification and more general in the atherogenic process has been invoked,
that includes the differentiation of vascular smooth muscle cells into chondrocytes
with a osteoblast-like phenotype [22,24]. Curiously enough, the increased risk persists even when
correcting for low serum 25OHD, while low serum 25OHD is also emerging as
independent risk factor for cardiovascular disease [25].
We purposed to evaluate mineral metabolism in diabetic patients
assisted at the outpatient endocrine department. We wanted to assess how common
abnormalities of mineral metabolism were in the real setting of diabetes care,
to explore if these abnormalities could help explain the increased fracture
risk despite an increased bone mineral density and to verify whether indexes of
mineral metabolism were associated with micro- or macrovascular disease.
Patients and Methods
We studied all DM assisted by one of us, at a tertiary public
center. The database regarding those patients includes – gender and actual age,
time since diagnosis, height and actual weight from which the body mass index
was computed [weight(kg)/height(m)2, BMI], and the last available glycated hemoglobin [HbA1c]; presence or absence
of retinopathy – no retinopathy, retinopathy without laser therapy and
retinopathy with previous laser therapy (summed up when indicated as
retinopathy yes or no), nephropathy - no nephropathy, microalbuminuria > 30
mg/24h and renal failure with serum creatinine > 1.5 mg/dL (summed up when
indicated as nephropathy yes or no), and peripheral neuropathy - yes or no on
clinical questioning, were also recorded as well as the presence or absence of
High Blood Pressure (HBP) and dyslipidemia; drugs being used for diabetes
mellitus, HBP, dyslipidemia and anti-platelet agents were recorded; a specific
inquiry regarding use of Drugs or Supplements (DS) that included
anti-osteoporotic drugs, calcium or vitamin D were also included.
Besides the usual care [26], serum calcium, phosphate, magnesium, Parathyroid Hormone (PTH)
and 25-hidroxyvitamin D (25OHD) were obtained at the last appointment in a
venous sample drawn in the morning.
Analytical measurements were performed at the Chemical Pathology
Department of the hospital using standardized methods; chemical colorimetric
methods for serum calcium, phosphate and magnesium and Chemiluminescence
Immunoassay (CLIA) or Radioimmunoassay (RIA) methods for serum parathyroid
hormone - Immulite 2500 (Diagnostics Product Corporation, Los Angeles,
California) - and 25-hydroxyvitamin D - 25-HydroxyvitaminD 125I RIA kit (Dia Sorin
Inc., Stillwater). Intra- and inter-Assay variation coefficients were always
below 10%. Units and reference values are as follows: Serum calcium - 8.6-10.2
mg/dL; serum phosphate - 2.4-5.1 mg/dL; serum magnesium - 1.3-2.7 mg/dL; serum
parathyroid hormone (PTH) - 14-72 pg/mL; serum 25-hydroxyvitamin D (25OHD) -
30-80 ng/mL, with values < 20 ng/mL defining deficiency, values between
20-29 ng/mL insufficiency and values > 100 ng/mL toxicity.
Clinical and analytical data were entered on a Statistical Package
for the Social Sciences (IBM SPSS 24th version) database and statistical analysis used the same program.
Results are presented as the mean ±
standard deviation or as the percent as appropriate. The normal distribution of
continuous variables was assessed with the Kolmogorov-Smirnov test and
non-normal distributed variables were log transformed prior to analysis;
however, when no differences regarding the non-transformed variables were
found, these are presented instead for the sake of simplicity. Comparison
between variables used the chi-squared test or Anova with the Least Significant
Difference (LSD) for pos hoc comparisons as appropriate and the relation
between continuous variables used multiple regression analysis. The limit of
significance is 0.05 [27,28].
Results
Patient Characteristics
Only patients with Type 1 Diabetes Mellitus (DM1) - 56 patients
(16%) or patients with Type 2 Diabetes Mellitus treated with insulin (DM2IT) -
164 patients (47%) or Type 2 Diabetes Mellitus treated with oral agents (DM2OT)
- 130 patients (37%) were included in the analysis.
Clinical characteristics of the patients are summarized in Table 1. In short, patients
were both sexes, old (DM2IT and DM2OT) or young (DM1) adults, with long
standing disease and overweight (DM2IT and DM2OT). Metabolic control was less
than desirable, and they presented chronic microvascular disease and
macrovascular risk factors mainly in those with long standing Type 2 Diabetes
(DM2IT). In both DM2IT and DM2OT groups more than 80% of the patients were
using drugs to control blood pressure and more than 60% were using both
hypolipidemic drugs and anti-platelet agents. Drugs to treat osteoporosis or
containing calcium or vitamin D were only marginally used.
Differences between groups are presented although they were not
used later in the analysis; instead individual variables were selected to
explore mineral metabolism.
Abnormal Mineral Metabolism
Results regarding common parameters of mineral metabolism are
presented across the three defined groups (Table 2). Again, differences regarding groups although presented are not
further used in the analysis.
Regarding minerals, available as routine, results outside the
reference range were mild and uncommon: 28 cases for calcium (8%); 14 cases for
phosphate (4%) and 7 cases for magnesium (2%). Only four patients presented
more than one abnormal result. Analysis of the individual patient records
revealed the 45 patients (13% of the total) to present: primary
hyperparathyroidism previously undiagnosed (3 patients), secondary
hyperparathyroidism in patients with diabetic nephropathy (12 patients),
isolated hypocalcemia (2 patients) and isolated hypophosphatemia (2 patients)
secondary to serious gastrointestinal disease, hypocalcemia secondary to
hypovitaminosis D (2 patients), hypercalcemia secondary to vitamin D
toxicity (1 patient), hypercalcemia secondary to neoplastic disease (1
patient), hypercalcemia due to familial hypocalciuric hypercalcemia (1
patient), and 14 patients with isolated mild abnormalities secondary to
diuretic use (loop diuretics or thiazide-like diuretics); in 7
patients no obvious reason for the very mild abnormalities of serum calcium,
phosphate or magnesium could be found.
Decreased 25OHD levels were very common with 85% presenting
deficient/insufficient vitamin D levels (< 30 ng/mL) and 55% presenting
deficient vitamin D levels (< 20 ng/ml); only one patient presented
increased 25OHD levels (> 100 ng/mL) suggesting vitamin D toxicity. As noted
most of these patients presented normal serum values of calcium, phosphate and
magnesium (see later).
Increased PTH values were less common (21% of the patients) with
more than half of these presenting values above 100 pg/mL (13%). Low PTH values
were only found in 4 patients and were borderline with normal serum calcium
levels.
Relation to Diabetes Characteristics
Mineral indexes were not significantly different across sex and
were not significantly related to age. Time since diagnosis was significantly
related to both serum calcium (parcial r=-2.043, p<0.05) and to serum PTH
(partial r=2.036, p<0.05). Actual BMI was significantly related to serum
calcium (r=0.105, p<0.05) the relation increasing when all other mineral
parameters were simultaneously considered (partial r=0.268, p<0.01).
Metabolic control evaluated by the last HbA1c, was significantly
related to serum calcium (r=-0.129, p<0.05) to serum phosphate (r=0.162,
p<0.01) to serum PTH (r=0.171, p<0.01) and to 25OHD (r=-0.146,
p<0.05); however, when all variables were simultaneously considered only
vitamin D remained significant (partial r=-0.156, p<0.05) (Figure 1). Significance persisted
after correcting for nephropathy.
Patients with retinopathy presented significantly higher levels
of serum PTH compared to patients without retinopathy - 89±140 pg/mL vs. 52±40
pg/mL, t=3.039, df=348, p<0.01 - with no differences regarding other indexes
of mineral metabolism. Significance remained after correction for nephropathy.
Differences regarding mineral metabolism indexes in patients with
or without nephropathy are presented in Table 3.
Magnesium levels were significantly lower in those patients with
peripheral neuropathy - 1.9±0.3
mg/dL vs. 2.0±0.3 mg/dL, t=3.174,
df=348, p<0.01 - with no differences regarding serum calcium, phosphate, PTH
or 25OHD. Again, the differences persisted after correcting for nephropathy.
PTH levels were significantly higher in those patients with HBP -
70±103 pg/mL vs. 45±30 pg/mL, t=-1.965,
df=348, p<0.05 – with no significant differences regarding serum calcium,
phosphate, magnesium or 25OH; the significance was lost when nephropathy was
taken into account; however, in a model that includes HBP, nephropathy and age,
HBP is again a significant factor for serum PTH levels - r= 0.378, p<0.01,
partial rs 0.155, 0.320 and -0.178, respectively, p<0.05 in every case.
There were no significant differences regarding indexes of mineral metabolism
in patients with or without dyslipidemia.
As noted in Table 2, type 2 diabetic patients using insulin
(DM2IT) presented significantly lower calcium levels regarding type 2 diabetic
patients not using insulin (DM2OT) - 9.5±0.4
mg/dL vs. 9.7±0.5 mg/dL, t=2.489,
df=348, p <0.05 - with no other significant differences regarding the
indexes of mineral metabolism; however, the difference was reduced to a trend
(p<0.1) when nephropathy was taken into account. Patients using
thiazide-like diuretics - most of them in fixed associations with angiotensin
converting enzyme inhibitors or angiotensin receptor antagonists - presented
significantly higher calcium levels – 9.7±0.5
mg/dL vs. 9.5±0.5 mg/dL, t=3.274,
df=348, p<0.05 - in comparison with patients taking other HBP drugs or no
drugs, the differences persisting after correction for nephropathy, with no
other significant differences regarding the indexes of mineral metabolism.
There were no significant differences regarding the same indexes between
patients using statins compared to those not using them, nor in relation to
patients using or not acetylsalicylic acid as an anti-platelet agent. The
number of patients using drugs or supplements with calcium, vitamin D or
anti-osteoporotic agents was too small and furthermore with contradictory
effects on mineral metabolism to allow any meaningful comparisons.
Relation Between Indexes of Mineral Metabolism
Neither calcium (KS z=1.510, p<0.05), nor phosphate (KS
z=1.341, p<0.06) nor magnesium (KS z=1.806, p<0.05) were normal
distributed; this non-normal distribution persisted even if abnormal values
were excluded. Also, calcium, phosphate and magnesium were independent
variables - there was only a small albeit significant direct relation between
serum phosphate and serum magnesium (r=0.166, p<0.05).
Serum PTH (KS z=4.787, p<0.01) and 25OHD (KS z=2.009,
p<0.01) were also not normally distributed. Serum PTH was inversely and
significantly related to calcium (r=-0.187, p<0.01) and directly and
significantly related to phosphate (r=0.338, p<0.01) and to magnesium
(r=0.227, p<0.01); when the three variables were simultaneously considered
(n=350, r=0.345, p<0.01) all three remained as significant variables (β =
-27.752, 17.037 and 66.031, p< 0.01, 0.05 and 0.01, respectively for
calcium, phosphate and magnesium as dependent variables of PTH). On the
contrary 25OHD was not significantly related to serum calcium, to serum
phosphate or to serum magnesium not even when all three variables were
simultaneously considered or when PTH was also simultaneously entered in the
regression analysis.
25OHD levels were inversely related to serum PTH (r=-0.157, p <
0.05) (Figure 2), suggesting that for
levels of 25OHD less than 20 ng/dL, serum PTH progressively increases. However
only 21% of those presenting deficient vitamin D levels (< 20 ng/mL) and
only 16% of those with insufficient vitamin D levels (20-29 ng/mL) presented
increased PTH levels. Therefore, most patients with decreased vitamin D values
(< 30 ng/mL) presented normal PTH levels (81%). Only 2 patients with low
25OHD levels (< 30 ng/mL) (<1%) presented low serum calcium levels.
On the other hand, 74% of those patients presenting with increased
PTH levels presented low levels of 25OHD (< 30 ng/mL) and nephropathy, with
18% presenting only low levels of 25OHD without nephropathy, 4% only
nephropathy with normal 25OH levels and 4% both normal levels of 25OHD and
normal renal function. In fact, only 35% of those patients with nephropathy
presented high PTH levels. Ninety-one % of the patients with nephropathy
presented low levels of 25OHD, compared with only 78% of those without
nephropathy χ2 = 9.49, df= 1,
p<0.01.
Discussion
Abnormal mineral metabolism may be expected in DM patients, since
this is a chronic systemic disease with widespread organ involvement and
multiple drug use, and mineral metabolism may therefore be altered at several
levels - intake, absorption, distribution and gastrointestinal and renal
excretion [1-4].
Data regarding mineral metabolism in DM patients may help explain
the apparent paradox of an increased fracture risk in both type 1 and type 2
diabetics, even when accounting for several confounding variables, despite an
increased bone mineral density in type 2 DM patients [5-7]. This apparent paradox
suggests decreased bone quality beyond bone mineral density [10-12].
Data regarding mineral metabolism may be important by itself since
several indexes of mineral metabolism, including serum PTH, 25OHD and
hyperphosphatemia are now emerging as powerful independent risk factors for
cardiovascular disease and even all-cause mortality [22-25].
The sample of diabetic patients considered is large enough and
representative of diabetic patients assisted in tertiary centers in the western
world, to allow for meaningful conclusions; furthermore, any conclusions should
apply to the real practice conditions of everyday life, since they are
significant for the relatively small number of patients considered. Patients
are middle aged or old adults, with long standing disease, metabolic control
less than what would be desirable, and with evidence for microvascular disease
and risk factors for macrovascular disease. Multiple drugs were simultaneously
being used in these patients (average 13 “Pills” a day, data not shown).
Abnormal mineral metabolism is indeed common in diabetic patients
- 45 patients or one in every 7-8 patients (13%) - even if abnormalities of
calcium, phosphate and magnesium are generally mild, isolated and for the most
part secondary to diabetic nephropathy - 12 patients - or to diuretic use - 14
patients. It should be noted that besides the recognized effects of thiazide
and loop diuretics, the first one also evident in this sample, insulin use was
associated with lower serum calcium levels in this sample. Other cases - 14
patients - can also be expected in the general population including undiagnosed
cases of primary hyperparathyroidism, serious gastrointestinal disease,
hypovitaminosis D, vitamin D toxicity, neoplasia and familial hypocalciuric
hypercalcemia [15,17,29-32].
Much more common however are the low levels of 25OHD, that are
indeed the “norm” in diabetic patients. These results agree well with other
larger scale studies recently reported [33]. Interestingly enough in that study, an inverse relation of serum
25OHD levels with metabolic control was also reported, such as the one found in
this sample [33]. That relation may
explain the strikingly high prevalence of deficient/insufficient vitamin D
found in diabetics. However, the relation may also be bidirectional since some
studies have suggested that normal levels of vitamin D may protect against the
development of type 2 diabetes mellitus as recently reviewed (34). Vitamin D
receptors have been described in pancreatic β cells and vitamin D
insufficiency/deficiency has been associated with impaired insulin secretion [34]. In fact, protective
effects have been described for vitamin D supplementation on what relates do
retinopathy, nephropathy and neuropathy [34].
The clinical relevance of this data is however controversial since
as also observed in this sample almost all patients present normal calcium
levels (>99%) and normal serum PTH levels (79%). Only 21% of those
presenting deficient vitamin D levels and 16% of those with insufficient
vitamin D levels present increased PTH levels.
As recommended, we measure total 25OHD (D2 and D3, bound and free)
not the active form 1,25-dihydroxivitamin D, only one measurement was
considered despite the well-known seasonal variability and the reference range
is controversial, even if theoretically justified by maximal intestinal calcium
absorption and lowest PTH levels [33-38]. Furthermore, no routine assay for the vitamin D binding protein
is currently available. Even so it is remarkable that 9% of the patients
present extremely low levels of 25OHD (< 10 ng/mL), generally believed to be
associated with rickets/osteomalacia [35-38].
Much remains to be known about this hormone with ubiquitary
receptors distribution and widespread biologic effects [39]. Recent studies
suggest, serum 25OHD levels to be an independent risk factor for cardio- and
cerebrovascular disease and mortality at levels probably lower than those
considered appropriate for normal bone health [25,34,35,39].
Almost a fourth of the sampled patients also present increased PTH
levels, that for the most part seem to depend on the combined effect of
decreased 25OHD and nephropathy. In fact, diabetic patients may unique combine
these two factors, in a way similar to those with CKD (40,41). However, in
diabetes mellitus low levels of 25OHD occur prior to the development of
nephropathy in apparent relation to metabolic control. Besides in this report
at least, nephropathy was also associated unexpectedly to low levels of 25OHD,
not only the theoretically expected decrease of 1,25-dihydroxyvitamin D.
Contrariwise to the decreased 25OHD levels, the increased PTH may
be indeed real, since PTH was significantly and independently related to serum
calcium, phosphate and magnesium, and these variables were not normally
distributed which suggests a certain degree of constraint.
It should be noted that increased PTH levels - in fact even within
the normal range - are now emerging as a powerful independent risk factor for
cardiovascular disease and all-cause mortality in the general population and in
diabetic patients, generally associated with increased vascular stiffness [23-25]. Also in this study,
higher PTH levels were found both in patients with HBP and in patients with
retinopathy compared to patients without those conditions and the significance
remained after correcting for nephropathy. PTH may therefore be a risk factor
for both macro- and microvascular disease. How can the abnormal findings now
reported help explain bone fragility in diabetic patients?
Since no major widespread changes of serum levels of calcium,
phosphate or magnesium were found, a major role for hormones regulating mineral
metabolism and bone turnover is suggested.
The relevance of the decreased 25OHD seems doubtful. Firstly,
because the clinical relevance of low 25OHD is by itself unclear [34-38]; secondly because
decreased 25OHD would be expected at least in line with classic cases of
rickets and osteomalacia to result in mineralization defects of the expanding
bone matrix that would result in decreased bone mineral density not increased
bone mineral density and would result in more “soft” deforming bones [35-39].
The relevance of increased PTH seems more probable. First because
the increased levels seem real, explained by both the nephropathy and the
decreased 25OHD levels, and to be significantly related to serum calcium,
phosphate and magnesium [15]. Secondly because on bone PTH initiates the cycle of
reabsorption-formation that remodels bone, and increased PTH levels may lead to
uncoupling of the process resulting in bone cysts and bone brown soft tumors
that are the typical bone changes in primary hyperparathyroidism not
osteoporosis per se [42-45]. This is also suggested by the therapeutic use of teriparatide in
the treatment of serious osteoporosis and lack of effectiveness in promoting
bone healing after a fracture [46,47].
BMD is generally extremely low in osteitis fibrosa cystica the
classic bone disease of primary hyperparathyroidism [42-45]. Classic osteitis
fibrosa cystica however is now a rare manifestation of primary
hyperparathyroidism that more commonly is an incidental analytical finding in
asymptomatic subjects [42-45]. In a recent study half of the patients with primary
hyperparathyroidism presented normal BMD scores although decreased Trabecular
Bone Score (TBS) [43]. Despite this a much-increased risk of bone fractures is found
both at the vertebral and non-vertebral bone sites [46,47].
Bone disease in diabetic patients resembles in several ways bone
disease of CKD, namely because it includes low levels of vitamin D (25OHD or
1,25-dyhydroxyvitamin D) and increased PTH, even if in diabetic patients low
levels of 25OHD in relation to poor metabolic control may precede nephropathy
for many years and nephropathy seems associated in diabetic patients with
decreased levels of both 25OHD and 1, 25-dihydrovitamin D [41,48,49]. In both cases chronic
renal disease also leads to increased levels of Fibroblast Growth Factor
(FGF-23) induced by hyperphosphatemia, that may further reduce 1,25-OH-vitamin
D, by inhibiting kidney 1-α-hydroxylase activity [50].
In End Stage Renal Disease (ESRD) and milder forms of CKD an
increased fracture risk is found, but data regarding BMD is also contradictory;
an apparent decreased BMD in CKD disappears after correcting for confounding
factors in the National Health and Nutrition Examination Survey, and the Kidney
Disease Improving Global Outcomes recently recommended against BMD testing in
CKD stages 3-5, since it does not predict fracture risk as it does in the
general population and does not predict the type or renal osteodystrophy [41,48,49]. In the Health, Aging,
and Body Composition study including 2754 old (70-79 years) well functioning
community-living subjects, BMD at the femoral neck and total hip were not
significantly different in patients with or without CKD [49].
Tentatively, therefore the increased PTH levels may contribute
both to the increased bone mineral density and to the increased porosity - more
brittle bones found in diabetic patients.
Of course, factors related to diabetes may be thought to explain
the distinctive bone disease of diabetes. Hyperglycemia per se, Advanced
Glycosylation Products (AGE), oxidative stress, insulin, IGF1, hormones of the
adipose tissue, cytokines of the inflammatory cells of the adipose tissue, etc.
All have been tentatively implicated, but none so far seems to explain the increased
fracture risk despite an increased bone mineral density [10-12].
In brief we found that in the real conditions of diabetic care,
abnormal mineral metabolism is common and should be considered mainly in
relation to diabetic nephropathy, diuretic use and also insulin use. Apparent
vitamin D deficiency or insufficiency is extremely common and is inversely
related to metabolic control. Increased PTH levels although less common seem to
depend on low 25OHD levels and nephropathy, significantly relate to serum
calcium, phosphate and magnesium, may tentatively explain the increased bone
mineral density and more porous brittle bone, and to be by itself a significant
factor for micro- and macrovascular disease.
Conclusion
Standardized clinical care of diabetic patients reveals common
abnormalities of mineral metabolism (13%) mainly hyper- or hypocalcemia in
relation to diabetic nephropathy, diuretic and insulin use that must be
specifically addressed. Low levels of 25OHD are extremely common (85%) and
although inversely related to metabolic control and even lower in patients with
diabetic nephropathy are of uncertain relevance.
Increased PTH levels, although less common (21%) may be more
relevant. Increased PTH levels are found in relation to diabetic nephropathy
and low 25OHD levels a unique combination in diabetic patients, significantly
relate to serum calcium, phosphate and magnesium, may explain the increased
bone mineral density and more porous brittle bone found in diabetic patients
and may be a significant factor not only for macro- but also microvascular
disease. Evaluation of mineral metabolism must be added to the standardized
care of diabetic patients.
Acknowledgement: Authors wish to thank Rheumatology Colleagues José Carlos Romeu and Inês Seixas, for the many
valuable discussions.
Competing Interests Statement: Authors state there is no conflict of interest
regarding the presented work.
Funding: This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Figure 1: Linear regression of serum 25OHD vs
HbA1c (r=-0.146, p<0.05).
Figure 2: Linear regression of serum 25OH vs
serum PTH (r=-0.157, p<0.05).
|
DM1 (56) |
DM2IT (164) |
DM2OT (130) |
Differences between Groups |
Sex (F/M) (%) |
51/49 |
44/56 |
40/60 |
NS |
Age (years) |
39±14a |
67±11b |
66±12b |
F(2,348)=128.5, p<0.01 |
Years since diagnosis |
21±14a |
19±11a |
15±10b |
F(2,348)=9.3, p<0.01 |
BMI (kg/m2) |
24.3±3.9a |
28.9±5.2b |
28.8±5.1b |
F(2,348)=16.7, p<0.01 |
HbA1c (%) |
8.5±1.7a |
8.4±1.6a |
7.3±1.3b |
F(2,348)=28.1,p<0.01 |
Retinopathy (yes) (%) |
45a |
56a |
21b |
χ2 = 36.6, df=2, p<0.01 |
Nephropathy (yes) (%) |
38a |
56b |
40a |
χ2 = 14.2, df=2, p<0.01 |
Neuropathy (yes) (%) |
16a |
38b |
33b |
χ2 = 18.6, df=2, p<0.01 |
HBP (yes) (%) |
25a |
85b |
78b |
χ2 = 88.1, df=2, p<0.01 |
Dyslipidemia (yes) (%) |
27a |
63b |
62b |
χ2 = 25.9, df=2, p<0.01 |
HBP drugs (yes) (%) |
35a |
91b |
81c |
χ2 = 84.2, df=2, p<0.01 |
Hypolipidemic drugs (yes) (%) |
29a |
62b |
61b |
χ2 = 20.4, df=2, p<0.01 |
Antiplatelet drugs (yes) (%) |
21a |
68b |
66b |
χ2 = 49.3, df=2, p<0.01 |
DS (yes) (%) |
5 |
11 |
8 |
NS |
*Significant different groups are presented with different indexes a,b,c |
Table 1: Clinical Characteristics.
DM1 (56) |
DM2IT (164) |
DM2OT (130) |
Differences between groups |
|
Calcium (mg/dl) |
9.5±0.5 [8.1-10.5] |
9.5±0.4a [8.3-10.5] |
9.7±0.5b [8.1-12.0] |
F(2,348) = 3.212, p < 0.05 |
Phosphorus (mg/dl) |
3.6±0.8a [1.9-6.4] |
3.5±0.6 [1.8-5.8] |
3.4±0.6b [1.4-5.2] |
F(2,348) = 3.248, p < 0.05 |
Magnesium (mg/dl) |
2.0±0.2a [1.6-2.9] |
1.9±0.3b [1.3-2.9] |
1.9±0.3b [0.9-2.9] |
F(2,348) = 5.678, p < 0.01 |
PTH (ng/dl) |
80±183 [11-1210] |
63±50 [9-363] |
54±44 [28-124] |
NS |
25OHD (ng/dl) |
21±10 [4-44] |
22±14 [3-113] |
20±10 [4-54] |
NS |
*Range is indicated within square brackets. Significant different groups are presented with different indexes a,b,c |
Table 2: Parameters of Mineral Metabolism.
No nephropathy (192) |
Proteinuria (98) |
Renal failure (60) |
Differences between groups |
|
Calcium (mg/dL) |
9.6±0.4a |
9.5±0.4b |
9.5±0.5b |
F(2,348) = 3.8, p<0.05 |
Phosphate (mg/dL) |
3.5±0.6a |
3.4±0.7a |
3.8±0.8b |
F(2,348) =5.4, p<0.01 |
Magnesium (mg/dL) |
1.9±0.2a |
1.9±0.3a |
2.1±0.3b |
F(2,348) = 12.3,p<0.01 |
PTH (pg/mL) |
47±29a |
58±45a |
151±211b |
F(2,348) = 18.6, p<0.01 |
25OHD (ng/mL) |
24±14a |
20±9 |
17±6b |
F(2,348) = 4.2, p<0.05 |
*Significant different groups are presented with different indexes a,b,c |
Table 3: Mineral Metabolism Across Diabetic Nephropathy.
- Polonsky KS, Burant CF (2016) Type 2 Diabetes Mellitus.
In: Melmed S, Polonsky KS, Larsen PE, Kronenberg HM (ed.). Williams
Textbook of Endocrinology. (13th edition) Elsevier, Philadelphia 31, 1386-1450.
- Cagliero E
(2016) Diabetes and long-term complications. In: Jameson JL, De Groot LJ,
de Kretser DM, Giudice LC (ed.). Endocrinology: Adult & Pediatric. (7th
edition) Elsevier Saunders, Philadelphia. 51, 898-906.
- Brownlee M, Aiello LP, Cooper ME, Vinik AI, Plutzky J,
et al. (2016) Complications of Diabetes Mellitus. In: Melmed S, Polonsky
KS, Larsen PE, Kronenberg HM (ed.). Williams Textbook of Endocrinology (13th
edition) Elsevier, Philadelphia 33, 1484-1581.
- Gilbert MP, Pratley RE (2015) The impact of diabetes
and diabetes medications on bone health. End Rev 36: 194-213.
- Schwartz AV, Sellmeyer DE, Ensrud KE, Cauley JA, Tabor
HK, et al. (2001) Older women with diabetes have an increased risk of
fracture: a prospective study. J Clin Endocrinol Metab 86: 32-38.
- Bonds DE, Larson JC, Scwartz AV, Strotmeyer ES, Robbins
J, et al. (2006) Risk of fracture in women with type 2 diabetes: the
women’s health initiative observational study. J Clin Endocrinol Meta 91:
3404-3410.
- Hough FS, Pierroz DD, Cooper C, Ferrari SL (2016) IOF
CSA Bone and Diabetes Working Group Mechanisms and evaluation of bone
fragility in type 1 diabetes mellitus. Eur J Endocrinology 174: R127-R138.
- Ishii S, Cauley JA, Crandall CJ, Srikanthan P,
Greendale GA, et al. (2012) Diabetes and femoral neck strength: findings
from the hip strength across the menopausal transition study. J Clin
Endocrinol Metab 97: 190-197.
- Muka T, Trajanoska K, Jong JCK, Oei L, Uitterlinden AG,
et al. (2015) The association between metabolic syndrome, bone mineral
density, hip bone geometry and fracture risk: the Rotterdam study. PLOS
one.
- Melton LJ, Riggs L, Leibson CL, Achenbach SJ, Camp JJ
(2008) A bone structural basis for fracture risk in diabetes. J Clin
Endocrinol Metab 93: 4804-4809.
- Yamamoto M (2015) Insights into bone fragility in
diabetes: the crucial role of bone quality on skeletal strength. Endocr J
62: 299-308.
- Rubin MR, Patsch JM (2016) Assessment of bone turnover
and bone quality in type 2 diabetic bone disease: current concepts and
future directions. Bone Res 4: 16001.
- Leslie WD, Aubry-Rozier B, Lamy O, Hans D; Manitoba Bone
Density Program (2013) TBS (Trabecular Bone Score) and diabetes-related
fracture risk. J Clin Endocrinol Metab 98: 602-609.
- Kim JH, Choi HJ, Ku EJ, Kim KM, Kim SW, et al. (2015)
Trabecular bone score as an indicator for skeletal deterioration in
diabetes. J Clin Endocrinol Metab 100: 475-482.
- Bringhurst FR, Demay MB, Kronenberg HM (2016) Hormones
and disorders of mineral metabolism. In: Melmed S, Polonsky KS, Larsen PR,
Kronenberg HM (ed.). Williams Textbook of Endocrinology. (13th edition),
Elsevier, Philadelphia 28: 1254-1322.
- Golden SH, Robinson KA, Salanha I, Anton I, Ladnson PW
(2009) Prevalence and incidence of endocrine and metabolic disorders in
the United States: a comprehensive review. J Clin Endocrinol Metab 94:
1853-1878.
- Khosia S
(2008) Hypercalcemia and hypocalcemia. In: Fauci AS, Braunwald E, Kasper
DL, Hauser SL, Longo DL, et al.(ed.). Harrison’ s Principles of Internal
Medicine. (17th edition) McGraw-Hill Companies, New York. 47: 285-286.
- Parkman HP, Yates KP, Hasler WL, Nguyan
L, Pasricha PJ; The NIDDK Gastroparesis Clinical Research Consortium
(GpCRC), et al. (2011) Dietary intake and nutritional deficiencies in
patients with diabetic or idiopathic gastroparesis. Gastroenterology 141:
486-498.
- ShahM, Chandalia M, Adams-Huet B, Brinkley LJ, Sakhaee
K (2009) Effect of a high-fiber diet compared with a moderate -fiber diet
on calcium and other mineral balances in subjects with type 2 diabetes.
Diabetes Care 32: 990-995.
- Ellam T, Fotheringham J, Wilkie ME, Francis SE, Chico
TJA (2014) Bone mineral metabolism parameters and urinary albumin
excretion in a representative US population sample. PLOS One.
- Bilezikiam JP, Watts NB, Usiskin K, Polidori D, Fung A,
et al. (2016) Evaluation of bone mineral density and bone biomarkers in
patients with type 2 diabetes treated with canaglifozin. J Clin Endocrinol
Metab 101: 44-51.
- Hagstrom E, Hellman P, Larsson TE, Ingelsson E,
Berglund L, et al. (2009) Plasma parathyroid hormone and the risk of
cardiovascular mortality in the community. Circulation 119: 2765-2771.
- Yang B, Lu C, Wu Q, Zhang J, Zhao H, et al. (2016)
Parathyroid hormone, cardiovascular and all-cause mortality: a
meta-analysis. Clin Chim Acta 455: 154-160.
- Rashid G, Bernheim J, Green J, Benchetrit S (2007)
Parathyroid hormone stimulates endothelial expression of atherosclerotic
parameters through protein kinase pathways. Am J Physiol Renal Physiol
292: F1215-F1218.
- El Hilali J, de Koning EJ, van Ballegooijen AJ, Lips P,
Sohl E, et al. (2016) Vitamin D, PTH and the risk of overall and disease-specific
mortality: results of the longitudinal aging study Amsterdam. J Steroid
Biochem Mol Biol 164: 386-394.
- American Diabetes Association. Standards of Medical
Care in Diabetes - 2016. Position Statement. Diabetes Care 39: S6-S106.
- Wonnacott
TH, Wonnacott RJ (1990) Introductory statistics. John Wiley & Sons. (5th
edition,) New York.
- Norman GR,
Streiner DL (1994) Biostatistics. The bare essentials, Mosby Year Book
Inc., St. Louis.
- Carroll R, Matfin G (2010) Endocrine and metabolic
emergencies: hypercalcaemia. Ther Adv Endocrinol Metabol 1: 225-234.
- Carroll R, Matfin G (2010) Endocrine and metabolic
emergencies: hypocalcaemia. Ther Adv Endocrinol Metabol 1: 29-33.
- Seigneux S, Courbebaisse M, Rutkowski JM, Wilhelm-Bals
A, Metzger M, et al. (2015) Proteinuria Increases Plasma Phosphate by
Altering Its Tubular Handling. J Am Soc Nephrol 26: 1608-1618.
- Pham P-C T, Pham P-MT, Pham SV, Miller JM, Pham P-TT
(2007) Hypomagnesemia in patients with type 2 diabetes. Clin J Am Soc
Nephrol 2: 366-373.
- Katrinaki M, Kampa M, Margioris A, Castanas E,
Malliaraki N (2016) Vitamin D levels in a large Mediterranean cohort:
reconsidering normal cut-off values. Hormones 15: 205-223.
- Alam U, Arul-Devah V, Javed S, Malik RA (2016) Vitamin
D and Diabetic Complications: True or False Prophet? Diabetes Ther 7:
11-26.
- Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM,
Hanley DA, et al. (2011) Evaluation, treatment, and prevention of Vitamin
D deficiency: An Endocrine Society clinical practice guideline. J Clin
Endocrinol Metab 96: 1911-1930.
- Balvers MGJ, Brouwer-Brolsma EM, Endenburg S, de Groot
LCPGM, Kok FJ, et al. (2015) Recommended intakes of vitamin D to optimize
health, associated circulating 25-hydroxyvitamin D concentrations, and
dosing regimens to treat deficiency: workshop report and overview of
current literature. J Nutr Sci 4: 1-8.
- Demay MB,
Krane SM (2016) Disorders of mineralization. In: Jameson JL, De Grot LJ,
de Kretser DM, Giudice LC, Grossman AB, et al.(ed.). Endocrinology - Adult
and Pediatric. (7th edition) Elsevier Saunders, Philadelphia. 71,
1230-1243.
- Manson JE, Brannon PM, Rosen CJ, Taylor CL (2016)
Vitamin D deficiency - Is There Really a Pandemic? N Engl J Med 375 19:
1817-1820.
- St-Arnaud R,
Jones G, Glorieux FH (2016) Genetic defects in vitamin D metabolism and
action. In: Jameson JL, De Grot LJ, de Kretser DM, Giudice LC, Grossman
AB, et al.(ed.). Endocrinology - Adult and Pediatric. (7th
edition), Elsevier Saunders, Philadelphia. 67, 1160-1172.
- Lips P (2001) Vitamin D deficiency and secondary
hyperparathyroidism in the elderly: consequences for bone loss and
fractures and therapeutic implications. Endocr Rev 22: 477-501.
- Lundquist AL, Nigwekar SU (2016) Optimal management of
bone mineral disorders in chronic kidney disease and ESRD. Curr Opin
Nephrol Hypertens 25: 120-126.
- Silverberg
SJ, Bilezikian JP (2016) Primary hyperparathyroidism. In: Jameson JL, De
Grot LJ, de Kretser DM, Giudice LC, Grossman AB, Melmed S, Potts Jr JT,
Weir GC (ed.). Endocrinology - Adult and Pediatric. (7th
edition), Elsevier Saunders, Philadelphia. 63, 1105-1124.
- Bandeira
L, Bilezikian J (2016) Primary hyperparathyroidism. F1000 Res 5.
- Bandeira F, Cusano NE, Silva BC, Cassibba S, Almeida
CB, et al. (2014) Bone disease in primary hyperparathyroidism. Arq Bras
Endocrinol Metab 58: 553-561.
- Costa AG, Bilezikian JP (2013) Bone turnover markers in
primary hyperparathyroidism. J Clin Densitom 16: 22-27.
- Reid IR (2015) Short-term and long-term effects of
osteoporosis therapies. Nat Rev Endocrinol 11: 418-428.
- Shi Z, Zhou H, Pan B, Lu L, Liu J, et al. (2016)
Effectiveness of teriparatide on fracture healing: a systematic review and
meta-analysis. PLoS One 12: e0168691.
- Perry KW,
Salusky IB (2016) Chronic kidney disease mineral and bone disorder. In:
Jameson JL, De Grot LJ, de Kretser DM, Giudice LC, Grossman AB, et al,
(ed.). Endocrinology - Adult and Pediatric. (7th
edition) Elsevier Saunders, Philadelphia. 70, 1214-1224.
- Yenchek RH, Ix JH, Shlipak MG, Bauer DC, Rianon NJ;
Health, Aging, and Body Composition Study et al. (2012) Bone mineral
density and fracture risk in older individuals with CKD. Clin J Am Soc
Nephrol 7: 1130-1136.
- Erben RG, Andrukhova O (2016) FGF23-Klotho signaling
axis in the kidney. Bone 100: 62-68.