Pediatric Radiology

, Volume 43, Issue 2, pp 140–151

Rickets: Part I

Authors

    • Department of Medical Imaging, Box 9Ann & Robert H. Lurie Children’s Hospital of Chicago
    • Department of RadiologyNorthwestern University Feinberg School of Medicine
  • Russell W. Chesney
    • Department of Pediatrics, Le Bonheur Children’s HospitalUniversity of Tennessee Health Science Center
Minisymposium

DOI: 10.1007/s00247-012-2532-x

Cite this article as:
Shore, R.M. & Chesney, R.W. Pediatr Radiol (2013) 43: 140. doi:10.1007/s00247-012-2532-x

Abstract

Rickets is characterized by impaired mineralization and ossification of the growth plates of growing children caused by a variety of disorders, the most frequent of which is nutritional deficiency of vitamin D. Despite ample knowledge of its etiology and the availability of cost-effective methods of preventing it, vitamin D deficiency rickets remains a significant problem in developing and developed countries. This two-part review covers the history, etiology, pathophysiology and clinical and radiographical findings of vitamin D deficiency rickets. Other less frequent causes of rickets and some of the disorders entering into the differential diagnoses of rickets are also considered. Controversial issues surrounding vitamin D deficiency include determination of what constitutes vitamin D sufficiency and the potential relationship between low levels of vitamin D metabolites in many individuals and unexplained fractures in infants.

Keywords

RicketsVitamin DChildrenBoneMetabolic bone diseaseNon-accidental trauma

Introduction

Rickets is a disorder of growth plate mineralization and ossification, and hence is unique to children and adolescents prior to skeletal maturity. Although many disorders can cause rickets, the majority of cases, both historically and presently, are caused by vitamin D deficiency, which is the major topic of this two-part review. Other causes of rickets and differential diagnoses are also considered, as are some of the controversies surrounding vitamin D deficiency.

Table 1 provides the organization of this review. Part I covers the history of rickets, as well as the basic principles of bone and mineral physiology and growth plate anatomy needed to understand its pathophysiology. Vitamin D deficiency is discussed, including controversies concerning the amount of vitamin D needed by humans. Part II covers more specific features of vitamin D deficiency rickets, including its pathophysiology, clinical features, pathoanatomy and radiographical findings. Other forms of rickets and differential diagnoses are also discussed. The potential relationship between low levels of vitamin D metabolites in many individuals and unexplained fractures in infants is then considered.
Table 1

Table of contents

Part I

Part II

History of rickets

Vitamin D deficiency rickets

Rickets and osteomalacia: definitions and basic pathophysiology

 Pathophysiology

Review of bone formation and mineralization

 Pathoanatomy and radiographic features

Mineral homeostasis

Other forms of rickets

Bone mineral metabolism

 Nutritional rickets variants

Vitamin D metabolism and function

 Non-nutritional calcipenic rickets

Vitamin D deficiency

 Phosphopenic rickets

 Definitions

Rickets in special circumstances

 Controversial issues regarding vitamin D requirements

 Renal osteodystrophy

 Etiology/epidemiology

 Rickets with osteopetrosis

Prevention

Differential diagnoses

 

 Hypophosphatasia

 

 Metaphyseal chondrodysplasias

 

Vitamin D deficiency/non-accidental trauma controversy

History

Several reviews have covered the history of rickets [113]. Rickets is usually considered to have emerged primarily during the industrial revolution in northern Europe, with a demographic shift to urban environments and exposure to ultraviolet light limited by indoor lifestyle, narrow streets and dense smog from coal burning. Although rickets peaked during the latter part of the 19th century, it has been well documented in England since the 17th century [10]. The initial medical description of rickets is attributed to Daniel Whistler, who as a medical student published “Inaugural medical disputation on the disease of English children which is popularly termed rickets” in 1645, likely a synthesis of the observations of others. However, as the title indicates, it was a disease already recognized by the general public, although not described in the medical literature. In fact, documentation of rickets can be traced back at least 11 years earlier to 1634, when it was listed as the cause of 14 of the 10,900 deaths summarized in the Annual Bill of Mortality for the City of London [10]. Subsequently Francis Glisson, leading a team of investigators, published “De Rachitide” in 1650, providing a more definitive delineation of rickets, expanding on its clinical description and adding its pathological basis with autopsy correlation. Interestingly, the children investigated by Glisson were not from smog-filled cities but rather from more rural regions, children who had to spend most of their time working indoors when wool spinning became a home-based industry [9]. The origin of the word “rickets” has remained elusive; it is possibly from the Old English “wrickken” meaning “to twist” and possibly a reference to spinal deformity with “rachitis” indicating spinal inflammation. Both Whistler and Glisson addressed this issue but were unable to account for it, even considering that rickets had been named after an apothecary who had treated the disease.

With progression of the industrial revolution, the incidence of rickets increased. By the turn of the 20th century it was widespread in the industrialized cities of Europe and the northern United States, with clinically obvious rickets in 60–80% of children and higher incidences in autopsy series, including Georg Schmorl’s 1909 study showing evidence of rickets in 96% of children younger than 18 months. During this time, many observations and studies eventually came together to define vitamin D deficiency, which seemed to have dietary and environmental causation. Although not always mentioned, several earlier observations in the mid-1800s by Armand Trousseau (best known for his sign of hypocalcemia) included that rickets was caused by combined nutritional deficiency and a sunless environment, that it became manifest during periods of rapid growth and could be successfully treated with cod liver oil, and that there were similarities between rickets in children and osteomalacia in the mature skeleton [7]. In addition to widespread recognition that cod liver oil was effective in preventing rickets, other fats were also found to have an anti-rachitic effect. In 1919, Edward Mellanby demonstrated that cod liver oil, milk and butter were effective in preventing rickets and concluded that the anti-rachitic factor was either vitamin A or had a similar distribution. Elmer McCollum then showed in 1922 that oxidation of cod liver oil destroyed its effectiveness in preventing xerophthalmia but not rickets, distinguishing the anti-rachitic factor from vitamin A and identifying it as a new vitamin, designated as vitamin D.

The role of sun exposure in the prevention and cure of rickets was first suggested by Jedrzej Sniadecki in 1822. In 1890, Theobald Palm also proposed the importance of sunlight based on the geographical distribution of rickets, being rare in tropical countries despite poverty and unclean living conditions. However, it was not until nearly 30 years later that the anti-rachitic effect of ultraviolet light was shown, with such evaluation stimulated by rampant rickets in central Europe, which had been starved of dietary sources of vitamin D in the wake of World War I. Using a mercury vapour lamp in 1919, Kurt Huldschinsky showed that even the non-exposed arm was cured, indicating a systemic effect. Subsequently Alfred Hess and Lester Unger demonstrated sunlight to be effective. Further controlled studies of sunlight and cod liver oil by Harriette Chick verified the effectiveness of each in the prevention of rickets, paving the way for the eventual demonstration that dietary vitamin D and the cutaneously photosynthesized substance were the same.

In 1924, Harry Steenbock and Archie Black, followed by Hess and Mildred Weinstock, showed that UV irradiation of various foods produced an anti-rachitic substance. This led to the ability to produce plentiful vitamin D, initially from yeast, enabling prevention of rickets. In the United States this was accomplished mostly by vitamin D supplementation of milk, with long-acting intramuscular injections used in Europe. These methods were remarkably effective in eliminating the first wave of rickets that engulfed the northern hemisphere in the 19th and early 20th centuries. However, smaller epidemics of rickets have subsequently re-emerged, as discussed under the section on etiology and epidemiology.

This brief historical review is limited to rickets, the discovery of vitamin D and initial success in its treatment and prevention. The history of vitamin D biochemistry is provided elsewhere [6].

Rickets and osteomalacia—definitions and basic pathophysiology

Although rickets and osteomalacia are both disorders of deficient mineralization of organic matrix, there are fundamental differences. Rickets is a disease of the physes (growth plates) characterized not only by deficient mineralization of cartilage and osteoid but also by retarded endochondral ossification, which causes excessive accumulation of physeal cartilage, growth failure and skeletal deformities [11, 14]. Failure of normal apoptosis of hypertrophic chondrocytes is now recognized as the key abnormality causing this ossification defect. The abnormalities of mineralization and ossification are caused by insufficient circulating levels of calcium and phosphate ions. Although mineralization depends on the calcium × phosphate (Ca × P) product, the defect in endochondral ossification more specifically results from hypophosphatemia, which impairs chondrocyte apoptosis by inhibition of the caspase-9-mediated mitochondrial pathway [15, 16].

As rickets is a disorder of open growth plates, it is seen only in children. In osteomalacia, an insufficient Ca × P product causes failure of normal mineralization of osteoid, laid down either at sites of bone turnover or by the periosteum in the process of membranous bone formation. These processes occur in both adults and children. Hence osteomalacia can be present at any age. It is an oversimplification to define rickets as a mineralization disorder of children and osteomalacia as the equivalent condition in adults. Rickets involves the physes, and osteomalacia involves other sites of bone formation, but the more important concept for understanding rickets is recognition that the abnormalities of mineral ion homeostasis lead to skeletal deformity by disrupting endochondral ossification rather than just causing deficient mineralization of cartilage and osteoid. Rickets and osteomalacia are bone disorders that can be present simultaneously in children with disorders of mineral metabolism. Distinguishing between these is not just a matter of definition. Parfitt [17] indicated that these processes might differ in their pathophysiology, temporal course and response to therapy, referring to the different processes of crystal deposition at the physes versus appositional bone formation. When chondrocyte apoptosis and its effect on endochondral bone formation are considered, the differences between rickets and osteomalacia are even more pronounced.

Rickets and osteomalacia result from insufficient mineral ion concentrations, which are caused by many disorders. These are usually divided into “calcipenic” and “phosphopenic” categories, with these designations referring to whether the initial defect results in insufficient calcium absorption or excessive phosphate excretion, respectively. It does not divide rickets into hypocalcemic and hypophosphatemic groups. Because of compensatory changes, even calcipenic disorders lead to hypophosphatemia, and hence this categorization is not inconsistent with the current concept that hypophosphatemia is the common pathway to all rickets [16].

Calcipenic causes include vitamin D abnormalities and calcium deficiency. The most common vitamin D abnormality is vitamin D deficiency, requiring insufficiency of both dietary vitamin D and sunlight. It can also be caused by malabsorption, and obesity is an increasingly common cause of vitamin D deficiency from sequestration of vitamin D in adipose tissue [18, 19]. Much less common vitamin D abnormalities include defects in vitamin D metabolism and end-organ responsiveness. Although uncommon in most parts of the world, dietary calcium deficiency can cause rickets. In vitamin D disorders and calcium deficiency, rickets results from insufficient intestinal absorption of calcium.

Phosphopenic rickets is mostly caused by hereditary or acquired disorders of renal tubular phosphate wasting, the most frequent of which is X-linked hypophosphatemia (XLH, also known as familial vitamin D resistant rickets).

Review of bone formation and mineralization

Understanding the pathophysiology of rickets requires a review of normal endochondral bone formation (EBF) and mineralization [20]. In EBF, the skeleton is pre-formed in cartilage and then replaced by bone. This begins with mesenchymal condensation followed by differentiation to chondrocytes that secrete extracellular matrix proteins to form cartilage. Conversion of cartilage to bone begins in the centrally located primary ossification center, which grows peripherally, progressively forming bone. Its leading edge is called the ossification front, although it is the chondrocytes ahead of the ossification front that pave the way for its advancement. Subsequently, secondary ossification centers, the epiphyses, develop at the ends of long bones. The primary and secondary ossification centers continue to enlarge and grow toward each other, leaving only a thin strip of growth cartilage between them and the growth plate, or physis. The process of endochondral ossification is reflected in the histology of the physis (Fig. 1). Farthest from the ossification front is the resting zone with relatively few and randomly distributed small round chondrocytes. Moving centrally, the chondrocytes proliferate more rapidly and become flattened and arranged in orderly columns, forming the proliferating zone. Next, they stop proliferating and enlarge, forming the hypertrophic zone. Hypertrophic differentiation is a key step in endochondral ossification, which is highly regulated by many physiological factors, the most important being Indian hedgehog (Ihh) and parathyroid hormone-related peptide (PTHrP), to maintain normal growth plate organization and function [21]. Hypertrophic chondrocytes then undergo terminal differentiation and mineralize the surrounding cartilage matrix, forming the zone of provisional calcification (ZPC). Mineralization of cartilage is believed to be required for subsequent resorption by osteoclasts/chondroclasts [22]. Mineralization is followed by apoptosis of terminally differentiated chondrocytes. This removes the chondrocytes from cartilage columns and promotes the ingrowth of marrow elements, osteoblasts, osteoclasts and vessels from the metaphysis into tunnels between bars of calcified cartilage [16]. Osteoclasts/chondroclasts then resorb much of the cartilage matrix, and osteoblasts deposit osteoid (bone matrix) on the scaffold of residual calcified cartilage, forming the primary spongiosa.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-012-2532-x/MediaObjects/247_2012_2532_Fig1_HTML.gif
Fig. 1

Diagram of the functional anatomy of the growth plate. Growth plate function is reflected in its anatomy, which temporally progresses from the reserve or resting (R) chondrocytes closest to the epiphysis to metaphyseal (M) trabecular bone. As reserve chondrocytes proliferate they form columns in the proliferative zone (P) and also secrete chondroid matrix. These chondrocytes eventually stop proliferating and enter hypertrophic differentiation, a step that is highly regulated to enable normal growth plate function. In the hypertrophic zone (H), chondrocytes enlarge and release matrix vesicles (not shown), leading to cartilage calcification (C). Terminally differentiated hypertrophic chondrocytes then undergo apoptosis. There is vascular ingrowth with arrival of chondroclasts/osteoclasts, which resorb much of the calcified cartilage, and osteoblasts, which lay down osteoid on the remaining scaffold of calcified cartilage, forming trabecular bone of the metaphysis (M) (Drawing by Kittie Yohe and modified from Anderson and Shapiro [20] and Wallis [74])

Many of the processes involved in EBF are regulated by Ihh, a locally acting factor produced by hypertrophic chondrocytes [23]. It promotes chondrocyte proliferation, osteoblastic differentiation of mesenchymal cells and the perichondrial cells surrounding the growth plate. The membranous bone produced by these osteoblasts forms the leading edge of the cortex. Formation of this peripheral membranous bone might precede mineralization of the adjacent ZPC, producing a small bone spur, the “bone bark”, which is a normal finding [12]. Along with vascular endothelial growth factor (VEGF), Ihh also promotes vascular ingrowth for establishing the primary spongiosa.

In the simpler process of membranous bone formation, osteoblasts differentiate directly from mesenchymal cells and secrete osteoid. This accounts for formation of the calvarium, some other flat bones, the cortex of long bones and the bone formation that accompanies bone turnover and remodelling.

Biological mineralization is a complex process, not explained simply by precipitation of crystals from a supersaturated solution [17, 24]. In addition to an adequate Ca × P product, other conditions are required. Initiation of crystal formation, a particularly difficult process, begins either in specialized matrix vesicles or along collagen fibrils that have undergone specific modification to permit mineralization.

Overview of mineral homeostasis, bone mineral metabolism and vitamin D

Mineral homeostasis refers to the maintenance of normal circulating levels of calcium and phosphate ions, which are required for a wide array of physiological functions [25, 26]. Bone mineral metabolism refers to the processes that deposit or resorb mineral from bone, with a proper balance of these processes needed to maintain a normally mineralized skeleton. Mineral homeostasis and bone mineral metabolism are conceptually different and in some conditions at odds with each other. The major organs involved in these processes include the parathyroid, kidney, intestine and bone.

Basic mineral homeostasis [2527]

The major organ charged with maintaining a normal circulating calcium concentration is the parathyroid, with calcium levels monitored by its calcium sensing receptor (CaSR). If the ionized calcium concentration falls below a certain level, the CaSR instructs the parathyroid to increase the synthesis and release of parathyroid hormone (PTH). PTH then directs metabolic effects in bone, kidney and, indirectly, the intestine to raise the calcium concentration back to normal. For bone, PTH mobilizes calcium and phosphate by increasing osteoclastic bone resorption. In the kidney, PTH increases renal tubular reabsorption of calcium, decreases reabsorption of phosphate and up-regulates the enzyme that produces the active form of vitamin D. Activated vitamin D in turn has many functions, the most important of which is to increase intestinal absorption of calcium. This system accounts for calcium homeostasis, but not phosphate. Although our understanding of phosphate homeostasis has lagged behind that of calcium, considerable advances during the last decade have involved recognition of “phosphatonins”, which cause renal tubular phosphate wasting [2830]. The main phosphatonin is fibroblast growth factor-23 (FGF-23), discussed in the section on phosphopenic rickets. The major factors involved in mineral homeostasis are summarized in Fig. 2 and Table 2.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-012-2532-x/MediaObjects/247_2012_2532_Fig2_HTML.gif
Fig. 2

Diagram of basic calcium homeostasis. Solid arrows positive effect, dashed arrows negative effect. Circulating ionized calcium, via the parathyroid calcium sensing receptor, inhibits PTH synthesis and secretion. Hence, low Ca++ increases PTH. PTH then promotes osteoclastic bone resorption to mobilize calcium and phosphate. In the kidney, PTH increases tubular calcium reabsorption and synthesis of calcitriol by 1-OHase. Calcitriol then acts on the intestine to increase calcium absorption. Calcium mobilized from bone and absorbed from the intestine then corrects the hypocalcemia that initially induced PTH. The conjoined arrows for the effects of calcitriol and PTH on bone resorption reflect the permissive effect of calcitriol. Although bone resorption is predominantly an effect of PTH, it also requires the presence of calcitriol

Table 2

Major regulators of mineral homeostasis

Factor

Major effects

Major regulating factors

PTH

(1) Bone resorption to mobilise calcium and phosphate

(1) Increased by low Ca++ concentration via CaSR

(2) Up-regulates 1-OHase to promote calcitriol synthesis

(2) Decreased by calcitriol

(3) Increases calcium reabsorption in distal tubule

(3) Decreased by FGF-23

(4) Decreases phosphate reabsorption in proximal tubule

 

Calcitriol

(1) Increases gut absorption of Ca and phosphate

(1) Increased by PTH

(2) Permissive for PTH effect on bone resorption

(2) Decreased by FGF-23

FGF-23

(1) Increases phosphate excretion by decreasing tubular reabsorption

(1) Increased by calcitriol

(2) Down-regulates 1-OHase to decrease calcitriol synthesis

(2) Increased by high dietary or circulating phosphate, although no phosphate sensing receptor found

PTH parathyroid hormone, calcitriol 1,25-dihydroxy-vitamin D, FGF-23 fibroblast growth factor-23, 1-OHase 25-hydroxy-vitamin D-1α-hydroxylase, CaSR parathyroid calcium sensing receptor

Vitamin D metabolism and function

Vitamin D metabolism and function has been the subject of many excellent reviews [6, 8, 14, 27, 31, 32]. There are two forms of vitamin D. Ergocalciferol (vitamin D2) is produced by UV irradiation of ergosterol from plants, including yeast. Cholecalciferol (vitamin D3) is photosynthesized in skin from 7-dehydrocholesterol upon exposure to ultraviolet B (UV-B, 290–315 nm). Both forms follow the same metabolic pathways, are equivalent in their physiological effects, and henceforth are discussed together as vitamin D [18]. Vitamin D has essentially no metabolic activity of its own. Rather it is considered a pro-hormone, requiring further modification to become biologically active. It initially undergoes 25-hydroxylation in the liver by one of many cytochrome P450 enzymes to form 25-hydroxy-vitamin D (25D, also known as calcidiol). Because hepatic 25-hydroxylation is not physiologically regulated, serum 25D levels reflect vitamin D availability. 25D is also the major circulating metabolite and is stable, with a half-life of 2–3 weeks. Hence, the circulating 25D level is considered to be the best indicator of vitamin D status. 25D is then 1α-hydroxylated in renal proximal tubular cells by 25-hydroxy-vitamin D-1α-hydroxylase (1-OHase) to the active substance 1,25-dihydroxy-vitamin D, henceforth called calcitriol, its joint pharmaceutical and trivial chemical designation. Renal production of calcitriol by 1-OHase is highly regulated to maintain mineral and bone homeostasis, with 1-OHase activity up-regulated by PTH and down-regulated by FGF-23. Calcitriol in turn suppresses PTH not only by increasing calcium but by sensitizing the CaSR to suppression by calcium. Calcitriol also stimulates FGF-23 synthesis and inhibits its own synthesis. The effects of calcium and phosphate on calcitriol are largely mediated by PTH and FGF-23. After calcitriol is produced in the kidney, it enters the circulation to act remotely on the intestine, bone and parathyroid to correct the physiological conditions that stimulated its production, thus fulfilling the requirements for an endocrine hormone. Calcitriol is also structurally and functionally similar to steroid hormones, with its effects brought about by regulating gene expression. Like other steroids, calcitriol binds to an intracellular receptor (the vitamin D receptor, VDR) which then forms a heterodimer with the retinoid X receptor (RXR) and binds to DNA, specifically the vitamin D response element in the promoter region of 200–2,000 genes, triggering gene expression [8, 14, 27].

There are three categories of vitamin D effects to consider. Foremost is the maintenance of normal mineral ion concentrations, needed not only for prevention of rickets and osteomalacia but for many cellular functions and neuromuscular transmission [8]. Next there are skeletal effects, which are independent of the effects on mineral ions. Finally there are extraskeletal effects of vitamin D.

For maintaining mineral ion levels, the single most important effect of calcitriol is promotion of calcium absorption by the intestine. With vitamin D deficiency, only 10–15% of dietary calcium is absorbed, compared with 30–40% with vitamin D sufficiency. The first mechanism demonstrated for increasing calcium absorption is promotion of the gene for the calcium binding protein calbindin, which transports calcium across the enterocyte. Additionally, calcitriol promotes the formation of calcium channels, permitting entry of calcium into the enterocyte and extrusion of calcium from the enterocyte into the circulation [27]. Although intestinal phosphate absorption is also enhanced, this effect is relatively unimportant. While calcium balance is regulated by gut absorption as controlled by calcitriol, phosphate is absorbed more efficiently even without calcitriol and phosphate balance is regulated by renal tubular reabsorption, controlled by PTH and FGF-23. Calcitriol is also permissive for the effect of PTH on promoting osteoclastic bone resorption to mobilize calcium and phosphate; although primarily an effect of PTH, some calcitriol must also be present. The shared responsibilities of PTH and vitamin D in mineral metabolism can be viewed in two ways. Temporally, PTH is the rapid responder to aberrations in the calcium concentration, whereas vitamin D is more concerned with long-term mineral balance [26]. Functionally, PTH controls the circulating calcium concentration, whereas vitamin D is responsible for maintaining a normally mineralized skeleton [14].

There are conflicting data regarding whether vitamin D metabolites have direct effects in preventing rickets and osteomalacia, or if this is entirely explained by their effects on mineral ion concentrations. Reports of healing of rickets with vitamin D prior to correction of the mineral ion abnormalities argue for some direct effect, and many effects of vitamin D should directly promote bone mineralization [33]. For example, formation of organic matrix with the capability of being mineralized depends on calcitriol produced locally by 1-OHase in osteoblasts [26]. Extrarenal production of calcitriol in embryonic growth plate chondrocytes might have a role in normal endochondral ossification, including promotion of vascular invasion and prevention of rachitic-like widening of the hypertrophic zone [34]. Calcitriol also stimulates production of the calcium binding proteins osteopontin and osteocalcin by osteoblasts [27]. Despite these indicators of direct skeletal effects of vitamin D, there are compelling data that its anti-rachitic effects are caused solely by normalization of mineral ion concentrations [14, 35]. In calcitriol-resistant rickets (Part II), absence of a functioning calcitriol receptor prevents vitamin D signalling and severe rickets develops soon after birth. However, the rickets can be completely healed by calcium administration, suggesting that maintenance of mineral ion concentrations alone is sufficient to account for the anti-rachitic effects of vitamin D.

Vitamin D deficiency and nutritional rickets

Definition of vitamin D deficiency and associated controversies

Vitamin D sufficiency refers to how well the physiological requirements for vitamin D are being met, whereas vitamin D status refers to how well-supplied an individual is with vitamin D. While it is agreed that serum 25D levels are the best indicator of vitamin D status, the relation between vitamin D status and vitamin D sufficiency, although a matter of considerable importance, is not resolved. The bases for determining what constitutes vitamin D sufficiency are intertwined with many of the major controversies surrounding vitamin D. These include what constitutes full suppression of rickets and osteomalacia, the contribution of vitamin D toward preventing other skeletal disorders such as osteoporosis and fractures, and the potential role of vitamin D in prevention of non-skeletal disorders.

In vitamin D deficiency rickets, serum 25D levels are less than 15 ng/ml, and often less than 5 ng/ml1 [11]. The 1997 Institute of Medicine (IOM) report defined vitamin D deficiency by 25D levels less than 11 ng/ml [36]. Subsequently, higher values have been used for defining vitamin D deficiency and sufficiency. Many leaders in vitamin D research, and the World Health Organization, designate 25D levels less than 20 ng/ml as deficiency, 20–29 ng/ml as insufficiency, and 30 ng/ml or higher as sufficiency [31, 32, 37]. Because “deficiency” and “insufficiency” both mean that sufficiency has not been met, the term “insufficiency” when used in the context of vitamin D appears to refer to levels that are high enough to prevent clinically evident rickets and osteomalacia but not high enough, some knowledgeable people believe, to satisfy other health benefits of vitamin D.

Determination of vitamin D sufficiency involves assessment of whether its physiological functions are being fulfilled. Its most well-established role is the prevention of rickets and osteomalacia, for which its single most important effect is promotion of intestinal absorption of calcium. If this fails, calcium levels fall and PTH increases, leading to phosphaturia, hypophosphatemia and rickets. Hence, more sensitive means to determine whether vitamin D sufficiency has been achieved would be to determine whether the effects of vitamin D on calcium absorption and PTH suppression are satisfied. Studies suggest that maximal calcium absorption is reached at 25D levels of 32 ng/ml and that maximal PTH suppression is not reached until 25D levels of 30–40 ng/ml [32, 38]. Autopsy data showing elevated osteoid volumes (the hallmark of osteomalacia) in half of those with 25D levels between 20 ng/ml and 30 ng/ml support the contention that 20 ng/ml is not sufficient for suppression of osteomalacia [39]. Largely based on these considerations, many have indicated the need for levels of at least 30 ng/ml, hence the designation of 20–29 ng/ml as insufficiency. When using 30 ng/ml as the cut-off for vitamin D sufficiency, the prevalence of vitamin D deficiency/insufficiency is quite high. Based on data from the 2004 National Health and Nutrition Examination Survey (NHANES III), Adams and Hewison [31] indicate an overall prevalence of vitamin D deficiency/insufficiency in the United States of 90% for pigmented races and approximately 75% for whites. They also note that this represents a near doubling of the prevalence of deficiency/insufficiency over 10 years, considered largely due to increasing obesity with sequestration of vitamin D in adipose tissue.

In addition to the prevention of recognizable rickets and osteomalacia, vitamin D might have other beneficial effects on overall bone health. Many adult studies have shown positive effects of vitamin D supplementation on bone mineral density (BMD) and high correlations between 25D levels and BMD [40]. The effects of vitamin D (with or without calcium) supplementation on fracture rates have been summarized in meta-analyses. The 2006 meta-analysis by Bischoff-Ferrari et al. [41] suggested that optimal prevention of fractures is associated with 25D levels of at least 40 ng/ml, with a beneficial effect seen only in trials using at least 700 IU of vitamin D per day.

There are also arguments for benefits of higher levels of 25D than those needed to prevent clinically evident rickets and osteomalacia based on extraskeletal effects of vitamin D. Calcitriol directly or indirectly affects the function of at least 200, and perhaps as many as 2,000, genes, including those involved in regulation of cellular proliferation, differentiation, apoptosis and angiogenesis [27, 32, 42]. The presence of vitamin D receptors in many tissues suggests that they are influenced by vitamin D. 1-OHase is also present in many extrarenal tissues. Calcitriol produced at these extrarenal sites acts locally and its synthesis is not subject to the physiological regulation that controls renal production of systemic calcitriol, suggesting that these effects are particularly dependent on circulating 25D levels [27, 43]. All of these considerations suggest widespread physiological roles of vitamin D. Correspondingly, there has been considerable work linking vitamin D deficiency to a large variety of non-skeletal disorders, including lower extremity function and risk of falling, many cancers, autoimmune diseases including multiple sclerosis and rheumatoid arthritis, diabetes mellitus types 1 and 2, susceptibility to infectious diseases including TB, influenza and upper respiratory tract infections, neurological disorders, hypertension, peripheral vascular disease, asthma and many other disorders [32, 42, 44]. Multiple lines of evidence support roles of vitamin D in the prevention of these disorders, including laboratory evidence of underlying mechanisms, epidemiological evidence, case-control studies of vitamin D levels and clinical trials of vitamin D supplementation. Of these, the effects of vitamin D on the immune system— particularly with respect to tuberculosis—and on colorectal cancer are briefly reviewed, as illustrative examples of the vast amount of work that has been done on extraskeletal effects of vitamin D, and to indicate the difficulty and lack of consensus in determining whether this work supports the need for vitamin D supplementation.

Historically, TB and rickets/vitamin D deficiency have been linked [45, 46]. During the industrial revolution when rickets was ubiquitous, TB was also rampant, leading to initial supposition that rickets was an infectious disorder that spread through crowded urban slums, although Robert Koch was unable to isolate an infectious agent. A specific association of rickets and TB was also recognized by Sir William Jenner in 1895. Prior to effective antimicrobial therapy, the benefits of “fresh air and sunlight” were also appreciated, including recognition that high-altitude sunlight, which enhanced UV-B exposure, was particularly beneficial, with the first high-altitude TB sanatorium established by Hermann Brehmer in 1854. Niels Finsen’s successful treatment of lupus vulgaris (cutaneous TB) with sunlight led to a Nobel Prize in 1903. More recently, case-control studies have shown a relationship between low 25D levels and TB as summarized in a meta-analysis [47].

Extensive evaluation of the effects of vitamin D on the immune system has shown an overall suppressive effect on the adaptive immune response, considered to account for its preventive effect against certain autoimmune disorders. However, it has a positive effect on innate immunity, which is believed to be important in protection against tuberculosis [27, 46] and might represent the primitive function of vitamin D, predating its endocrine role in mineral metabolism [31]. In 1986, Rook et al. [48] showed that calcitriol promotes intracellular killing of TB-infected macrophages in vitro. Subsequent work has shown that intracellular killing involves antimicrobial peptides, the most important being cathelicidin, and that their synthesis depends on calcitriol produced locally by macrophages. Along with these observational and laboratory indicators of a role of vitamin D in protection against TB, controlled clinical trials of vitamin D added to standard antibiotic therapy have shown beneficial effects in both children and adults, although the data are limited [49, 50].

Multiple lines of evidence also support a protective effect of vitamin D against colorectal cancer. The initial link was based on disease incidences at varying distances from the equator, and hence incident sunlight [51]. Numerous studies have demonstrated an association between colorectal cancer and low serum 25D levels [52]. Extensive laboratory work supporting a protective role of vitamin D includes the ability of calcitriol to decrease proliferation and increase differentiation of colorectal cancer cells, and effects on proto-oncogenes, growth factor-activated pathways and detoxification of carcinogens [53]. There is also an inverse correlation between 25D levels and the proliferative component of the colonic mucosa. Despite these observational and experimental suggestions that vitamin D should be helpful in preventing colorectal cancer, whether there is a beneficial effect of vitamin D supplementation, as supported by randomized clinical trials (RCTs), is not clear. Davis and Milner [54] have concluded that the available RCTs have failed to show a beneficial effect, and they further warn that excessive vitamin D exposure might be a risk factor for cancer of the oesophagus, pancreas and prostate. However, Bischoff-Ferrari et al. [41] claim that if analysis of the available studies is limited to subjects receiving sufficient vitamin D, then a beneficial effect is demonstrated. It is also argued that the long latencies of many cancers make them not amenable to evaluation by RCTs, and that the observational and mechanistic data should be sufficient [55].

Despite the vast amount of work on the skeletal and extraskeletal beneficial effects of vitamin D, much of which appears quite compelling, the 2011 IOM report did not endorse the need for 25D levels greater than 20 ng/ml [56]. Furthermore, recognizing individual variability in the amount of vitamin D needed to support physiological function, the IOM indicated that its use of 20 ng/ml as a laboratory cut-off for vitamin D deficiency meant that it meets the needs of 97% of individuals, with 16 ng/ml meeting the needs of 50% and 12 ng/ml meeting the needs of 3%. IOM also specifically indicated that presently there was insufficient RCT data for recommending higher vitamin D levels based on potential extraskeletal effects of vitamin D. So different were these conclusions from those of many researchers in vitamin D that just prior to the release of the 2011 3rd edition of Vitamin D (editor D. Feldman), all of the authors were asked to consider revising their chapters upon consideration of that report, although many were not dissuaded from their belief in the importance of vitamin D for prevention of many diseases [57]. The difference between the IOM definition of vitamin D adequacy and 30 ng/ml, as recommended by many, is huge. It changes whether the majority of the United States population is considered to have sufficient vitamin D status versus vitamin D insufficiency that requires supplementation, with obvious implications for public health policy. If benefits of higher vitamin D levels exist, we would be remiss in not recommending appropriate supplementation. However, to recommend this for the entire population in the absence of supporting data is not desirable, either from the standpoint of allocation of healthcare resources or a scientific approach to the practice of medicine.

Why is the amount of vitamin D needed by humans unresolved and what is the basis for disagreement? There was no shortage of scientific expertise or diligence on the part of the IOM panel, nor on the part of many who disagree with its conclusions. It is likely that the major factors contributing to this lack of consensus are the coexistence of such a large volume of data to be reviewed, only a small portion of it coming from RCTs, and the premise by the IOM that its recommendations should be based largely on RCTs. The non-applicability of RCTs for some of the potential long-term health benefits of vitamin D is an issue [55]. An argument has also been made concerning where the burden of proof ought to lie when the data are not conclusive. For drugs with pharmacological effects, no beneficial effect should be assumed until proved. However, for a physiological substance such as vitamin D, different rules may be more appropriate. Based on estimates of sunlight exposure and latitude, it has been suggested that during much of human evolution enough vitamin D was photosynthesized to maintain 25D levels of 40–80 ng/ml, which should be considered to be the natural state. While this does not prove that these levels are needed, it is argued that it might be more appropriate for the burden of proof to lie with those who propose that lower levels are sufficient [58].

Expert consensus on the 25D levels needed for vitamin D sufficiency remains a work in progress. All agree that levels of at least 20 ng/ml should be maintained, and hence the designation of deficiency for lower levels, recognizing that not everyone below 20 ng/ml is actually physiologically deficient. Although the IOM specifically indicates that levels of at least 20 ng/ml are considered normal, the subsequent 2011 clinical practice guidelines of The Endocrine Society maintain a designation of insufficiency for 21–29 ng/ml [59]. Of note, consideration of 25D levels between 21 ng/ml and 29 ng/ml as insufficient has not gained official acceptance for children in the United States as reflected by guidelines of the Committee on Nutrition of the American Academy of Pediatrics [60] and the Pediatric Endocrine Society [61], likely related to a greater emphasis on the anti-rachitic effects of vitamin D by these paediatric organizations.

Etiology and epidemiology of vitamin D deficiency rickets

Most foods, other than oily fish, contain very little vitamin D unless artificially supplemented. Cutaneous synthesis of vitamin D is influenced by multiple factors, with effective UV-B exposure decreased by latitude from the equator, winter months, sunscreens and skin pigmentation. UV-B is absorbed by melanin and darker-skinned persons require 5–10 times the amount of sun exposure for equivalent vitamin D production [62]. Races evolving distant from the equator are usually fair-skinned, maximizing vitamin D synthesis from limited UV-B, other than the darker-skinned arctic Inuit population whose diet is rich in oily fish containing vitamin D [63]. This suggests that over the course of human evolution the major source of vitamin D has been cutaneous photosynthesis, with little contribution from natural food. Presently vitamin D fortification in the United States includes milk and milk products, as well as some cereals and breads. However, the vitamin D supplied by these sources is limited; sunlight exposure remains important in maintaining vitamin D sufficiency and factors limiting cutaneous photosynthesis play a substantial role in the etiology of vitamin D deficiency rickets.

Because vitamin D deficiency rickets is particularly problematic in breast-fed infants, attention has been given to those aspects of vitamin D related to the mother-infant pair [19, 64, 65]. During pregnancy, there is free placental transfer of 25D, but neither vitamin D nor calcitriol. The fetal requirement for vitamin D is limited as patients with essentially no vitamin D signalling appear normal at birth, although this does not exclude subtle effects of maternal vitamin D deficiency. Relatively recently, low maternal 25D levels were shown to correlate with increased fetal distal femoral splaying, determined by US measurements of femoral length and metaphyseal width [66]. It has been suggested that these findings are equivalent to rickets, possibly resulting from deficient production of calcitriol by fetal chondrocytes [67]. With maternal vitamin D sufficiency, the fetus receives enough 25D for short-term postnatal adequacy, limited by its half-life of 2–3 weeks. If additional vitamin D is not supplied or synthesized, the infant becomes vitamin D deficient—more quickly for those beginning with low levels. Following birth, the pattern of transfer of vitamin D from mother to infant changes, with breast milk containing 20–30% of the maternal vitamin D and essentially no 25D or calcitriol. Hence, the supply of vitamin D in breast milk is limited, particularly if maternal vitamin D levels are low. A high association between low maternal 25D levels and vitamin D deficiency in breast-fed infants has been demonstrated in multiple populations worldwide, as summarized by Thandrayen and Pettifor [19]. This reflects a combination of insufficient vitamin D in breast milk from these mothers as well as exposure of the infant to the same environment as the mother, with limited sunlight exposure.

The epidemic of rickets linked to the industrial revolution that was most prominent in the late 19th and early 20th centuries constituted the first wave of rickets. This was largely eliminated following the discovery of vitamin D and food fortification. However, rickets re-emerged. The second wave developed during the 1960s to 1980s in breast-fed infants from distinct cultural groups, relatively limited in overall numbers. These included infants born to dark-skinned Asian women who had immigrated to England, Muslim women in the Middle East and Asia, and members of certain religious sects in inner cities in the United States, all of whom had cultural practices, including extensive clothing, that limited sunlight exposure. With the promotion of breast-feeding, the third wave of rickets then emerged in the 1990s among breast-fed infants of dark-skinned American women, mostly African Americans, and this continues as an ongoing public health problem [3]. Unlike the second wave, these third-wave mothers are mostly normally dressed and part of mainstream society. However, vitamin D photosynthesis is limited for these mothers and their infants by skin pigmentation and relatively limited sun exposure in urban environments, a result of multiple factors that prevent them from going outdoors, including fear of community violence, work schedules and warnings against the dangers of sun exposure [68]. The overall effects of race on vitamin D deficiency in infants in the northern United States are highlighted in data from Pittsburgh showing low 25D levels in 45.6% of African-American versus 9.7% of white neonates, with skin pigmentation, socio-economic factors and cultural practices all considered to contribute to vitamin D deficiency in African Americans [69].

Although breast-fed infants are at greatest risk, significant vitamin D deficiency remains problematic throughout childhood and adolescence [70]. Adolescents with vitamin D deficiency might present with hypocalcemia rather than classic rickets, thought to be related to increased demands for calcium during periods of rapid growth, similar to early infancy [71].

Prevention of vitamin D deficiency rickets

Despite adequate knowledge of its etiology, vitamin D deficiency rickets remains a significant problem. The 1997 IOM report [36], endorsed by the American Academy of Pediatrics (AAP) in 2003 [72], suggested that infants, children and adolescents receive 200 IU of vitamin D beginning in the first 2 months of life. That recommendation was based on data showing that 200 IU prevented overt clinical manifestations of rickets and maintained 25D levels of at least 11 ng/ml. With the recognition that higher levels of 25D are needed, these recommendations have been revised. Based on maintaining 25D levels of at least 20 ng/ml, the 2008 AAP report recommends that all infants, children and adolescents receive 400 IU orally, either through diet or supplementation, beginning soon after birth [60]. Alternatively, with high-dose maternal supplementation, breast milk can supply sufficient vitamin D for the prevention of rickets [73], although the AAP stresses that the safety of this approach is not well established. Similar to the 2008 AAP recommendations, the 2011 IOM report also recommends 400 IU/day during the first year of life [56]. Beyond the first year of life the IOM “recommended dietary allowance” is 600 IU/day to age 70 and 800 IU beyond age 70. However, if the goal of vitamin D supplementation is achievement of 25D levels of 30 ng/ml, daily doses of 1,500–2,000 IU/day would be required [42, 59].

The proper role of sun exposure in meeting our vitamin D requirement is also controversial. Although sunlight has been the major source of vitamin D, recognition of its role in causing skin cancer has led to a strong public health awareness campaign on the dangers of sunlight, initially sponsored by the American Academy of Dermatology. Some now regard any unprotected sun exposure as too risky and suggest that vitamin D requirements be entirely met orally. However, reliance on oral vitamin D is problematic. As there are relatively few natural sources of vitamin D, this usually requires the use of supplements. Because of adherence issues, recommendations for prevention of vitamin D deficiency that are based on the use of supplements in lieu of sun exposure have a risk of inadequate intake [60]. Although the 2011 IOM report did not consider a role for sun exposure, without it the amounts of vitamin D recommended by the IOM would not achieve vitamin D sufficiency [58]. Recognizing clear risks of excessive sun exposure, the current trend towards strict sun avoidance and use of sunscreens might be over-zealous. Using sunlight in moderation would help provide for more reliable prevention of vitamin D deficiency, and this might be considered by the IOM in the future.

Summary

Part I of this review emphasizes the pathophysiological processes that underlie rickets. Rickets is a disorder of the growth plates of children in which impaired mineralization and endochondral ossification result from deficient mineral ion concentrations. The ossification defect is caused by an arrest of apoptosis of hypertrophic chondrocytes.

The most common etiology of rickets, historically and presently, is vitamin D deficiency. Accordingly, vitamin D metabolism and function and the epidemiology of vitamin D deficiency have been reviewed. Controversial issues related to the definition of vitamin D deficiency and its prevention have also been discussed.

Footnotes
1

As is customary in the United States and Canada, 25D levels are expressed in ng/ml. However, there is a trend toward using nmol/l; 1 ng/ml = 2.5 nmol/l

 

Conflicts of interest

None.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012