Article Text

Download PDFPDF

Nutritional rickets under 16 years: UK surveillance results
  1. Priscilla Julies1,
  2. Richard M Lynn2,3,
  3. Karina Pall4,
  4. Marina Leoni4,
  5. Alistair Calder5,
  6. Zulf Mughal6,
  7. Nick Shaw7,8,
  8. Ciara McDonnell9,
  9. Helen McDevitt10,11,
  10. Mitch Blair12
  1. 1 Child Health, Royal Free London NHS Foundation Trust, London, UK
  2. 2 Institute of Child Health, University College London Research Department of Epidemiology and Public Health, London, UK
  3. 3 BPSU, Royal College of Paedaitrics, London, UK
  4. 4 BPSU, Royal College of Paediatrics and Child Health, London, UK
  5. 5 Department of Radiology, Great Ormond Street Hospital, London, UK
  6. 6 Department of Paediatric Endocrinology, Royal Manchester Children's Hospital, Manchester, UK
  7. 7 Endocrinology and Diabetes, Birmingham Women's and Children's NHS Foundation Trust, Birmingham, West Midlands, UK
  8. 8 Institute of Metabolism and Systems Research, University of Birmingham, Birmingham, West Midlands, UK
  9. 9 Endocrinology Department, Children's University Hospital, Temple Street, Dublin, Ireland
  10. 10 Paediatric Bone and Endocrinology, Royal Hospital for Children Glasgow, Glasgow, UK
  11. 11 Neonatology, Royal Hospital for Children Glasgow, Glasgow, UK
  12. 12 Paediatrics, Imperial College London, Harrow, UK
  1. Correspondence to Dr Priscilla Julies, Child Health, Royal Free London NHS Foundation Trust, London NW3 2QG, UK; p.julies{at}nhs.net

Abstract

Objective The UK national incidence of nutritional rickets is unknown. We aimed to describe the incidence, presentation and clinical management of children under 16 years with nutritional rickets in the UK presenting to secondary care.

Methods Prospective data were collected monthly between March 2015 and March 2017 from 3500 consultant paediatricians using British Paediatric Surveillance Unit methodology. Clinicians completed online clinical questionnaires for cases fitting the surveillance case definition.

Results 125 cases met the case definition, an annual incidence of 0.48 (95% CI 0.37 to 0.62) per 100 000 children under 16 years. 116 children were under 5 years (annual incidence of 1.39 (95% CI 1.05 to 1.81) per 100 000. Boys (70%) were significantly more affected than girls (30%) (OR 2.17, 95% CI 1.25 to 3.78). The majority were of Black (43%) or South Asian (38%) ethnicity. 77.6% of children were not taking vitamin D supplements despite being eligible. Complications included delayed gross motor development (26.4%), fractures (9.6%), hypocalcaemic seizures (8%) and dilated cardiomyopathy (3%). Two children died (1.6%). In eight cases, rickets was confirmed radiologically and biochemically [raised serum alkaline phosphatase (ALP) and parathyroid hormone (PTH) levels ] but were excluded from the incidence analysis for not meeting the case definition of 25-hydroxyvitamin D of <25 nmol/L.

Conclusion The incidence of nutritional rickets in the UK is lower than expected. Serious complications and unexpected deaths, particularly in Black and South Asian children under 5 years, occurred. Both vitamin D deficiency and dietary calcium deficiency are role players in pathogenesis. Uptake of vitamin D supplementation remains low.

  • paediatric practice
  • nutrition
  • bone disease

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic?

  • The apparent re-emergence of rickets in children in many countries, including the UK, has been of concern around the adequacy of public health preventive measures. No recent UK-wide surveillance study has been carried out for nutritional rickets.

What this study adds?

  • The incidence of nutritional rickets presenting to secondary care is low: 0.48 (95% CI 0.37 to 0.62) cases per 100 000 children under 16 years.

  • Supplementation is still suboptimal in the majority of cases.

Implications of all available evidence

  • UK-wide surveillance has indicated the persistence of nutritional rickets with serious clinical consequences in a few.

  • Further attention is needed on what and how public health messages can target and reinforce vitamin D supplementation and sun exposure policy.

  • A wider surveillance research definition and the inclusion of additional sources, such as laboratory data and primary care and other physicians, may increase more precise ascertainment in the future.

Introduction

Nutritional rickets is a disorder of growing children, arising from deficiency of vitamin D, dietary calcium or both, resulting in impaired mineralisation of the growth plate and osteoid. Previously common, it was considerably reduced in the UK and the USA by food fortification, cod liver oil supplementation and environmental changes.1 It has recently re-emerged as a major public health concern2 worldwide with multiple risk factors. Calcium-deficiency rickets is described as occurring more in low-income countries with differing risk factors but also exists in high-income countries.3

While surveys and observational studies4 5 describe the re-emergence of symptomatic vitamin D deficiency in the UK6 and vitamin D deficiency hypocalcaemic seizures,7 the exact incidence of nutritional rickets in the UK is unknown.

We aimed to describe the incidence, presentation and clinical management of children with nutritional rickets in the UK presenting to secondary care to provide current, robust data to inform public health policy.

Methods

Standard British Paediatric Surveillance Unit (BPSU) surveillance methodology8 was used; for 25 months from March 2015 to March 2017, 3500 consultant paediatricians receiving a monthly electronic card (e-card) reported cases fitting the surveillance case definition (box 1). Reporting clinicians completed an online clinical questionnaire, providing data on demographics, known risk factors, clinical presentation, referral and treatment options used for these patients.

Box 1

Surveillance case definition

Clinical rickets:

  • Leg deformity/swollen wrists, knees or ribs and 25-hydroxyvitamin D (25-OHD) of <25 nmol/L with one or more abnormalities of serum calcium, alkaline phosphatase, phosphate and parathyroid hormone.

or

radiological rickets:

  • Widening, cupping and splaying of metaphysis (of any long bone) and 25-OHD of <25nmol/L.

Clinicians exported radiographs via hospitals’ image exchange portal using the project reference number. An expert radiologist determined whether rickets was present or not. A general paediatrician, project manager and, subsequently, the project team reviewed clinical cases for inclusion, using the strict case definition.

Exclusion criteria included (1) genetic rickets, (2) rickets associated with other chronic diseases and (3) metabolic bone disease of prematurity (infants whose corrected age at presentation is <3 months, born at <36 weeks’ gestation and weighing <1.5 kg).

Statistical analysis was performed using IBM SPSS Statistics for Windows V.25.0 and Microsoft Excel V.2016. Descriptive analysis was performed using t-test and χ2 tests.

Results

Ascertainment

E-card response compliance during the study period was 94.2%. One hundred and twenty-five cases were included for analysis. Eighty-four cases from completed questionnaires were excluded as they did not meet the case definition, were outside of the reporting period and notified in error or duplicates. Thirty-five more cases were defined as ‘not meeting the surveillance definition’ and are described later (figure 1).

Figure 1

Ascertainment of cases reported to the British Paediatric Surveillance Unit.

An overall incidence of 0·48 (95% CI 0.37 to 0.62) cases per 100 000 children under 16 years was calculated using the surveillance case definition and UK Office of National Statistics population estimates.9

London, East Midlands, West Midlands, Yorkshire, The Humber and Scotland had the highest incidence rates (figure 2).

Figure 2

Estimated annual incidence per 100 000 children (95% CI) by region.

Demographics

Boys (n=87, 70%) were significantly more affected than girls (n=38, 30%) (OR 2.17, 95% CI 1.25 to 3.78, p=0.005), and children under 5 years (n=116), aged 12–23 months, were the most affected age group (table 1).

Table 1

Age at presentation

The median age at diagnosis was 18 months (IQR 11.25–24.0, range 2–181). Excluding missing data, most cases were of Black (n=54, 44%) or South Asian (n=47, 39%) ethnicity (table 2).

Table 2

Ethnicity

Most cases presented in the spring (41/125, 32.8%), summer (29/125, 23.2%) and winter (28/125, 22.4%), but no statistically significant seasonal variability occurred (Pearson χ2 p>0.05).

Feeding practices

Eighty-five children were on solids, including dairy (57/85, 67.1%). Twenty-two (17.6%) children were exclusively breast feeding, but the mean duration was not calculable due to missing data. One child was exclusively formula fed at the time of diagnosis and presented at the age of 26 months with florid rickets.

Vitamin D supplementation

At diagnosis, 97 (77.6%) children were not reportedly receiving vitamin supplements (for eight cases, this information was not known or not recorded). Nutritional rickets occurred in 20 children (16%) who were on vitamin D supplements. Only three of these 20 children (n=3, 15%) were taking the government-recommended Healthy Start vitamins. Other supplements (n=14, 70%) used included Abidec (n=2, 10%), Dalivit (n=2, 10%) and unspecified preparations (n=10, 50%). Two children were on alfacalcidol. In one case, information was not recorded.

Maternal vitamin D status was unknown in most cases (n=83, 66.4%), not recorded (n=1, 0.8%) and, where known (n=41, 32.8%), 10 (8%) had vitamin D deficiency.

Associated conditions

Co-occurring conditions were described in 49 (39.2%) children. Cow’s milk protein allergy/intolerance (n=10, 8%), iron deficiency (n=7, 5.6%), eczema (n=7, 5.6%) and respiratory illnesses (n=7, 5.6%) were the most frequent. One child had a condition (gastrointestinal disease with fat malabsorption), and one child was taking a drug (glucocorticosteroid) known to be associated with vitamin D deficiency.10 In 33 (26.4%) cases, other conditions considered prior to rickets diagnosis were orthopaedic (n=8), neurological conditions (n=5), febrile seizures (n=3) and physiological bowing of legs (n=3).

Clinical presentation

Presenting symptoms ranged from limp, gait disturbances to acute life-threatening events, generalised seizures and safeguarding concerns. The most common clinical findings in confirmed cases were radiological features (n=91, 72.8%), bowed legs (n=61, 48.8%) and swollen wrists (n=50, 40%).

Complications

Complications included delayed motor development (n=33, 26.4%), fractures (n=12, 9.6%), hypocalcaemic seizures (n=10, 8%) and dilated cardiomyopathy (n=4, 3.2%), of whom two died (n=2, 1.6%).

Fractures occurred in seven boys and in five girls, aged 6–23 months and mostly identified in the femur (4/12, 33.3%). Two children were not mobile (table 3).

Table 3

Descriptive features of fractures in nutritional rickets

Hypocalcaemic seizures and rickets were reported only in boys, between 5 and 21 months. They were not receiving vitamin D supplements and exclusive breast feeding (n=7, 70%) was an additional risk factor. Febrile seizures, sepsis and epilepsy were considered prior to hypocalcaemia being recognised.

Three of the four children with dilated cardiomyopathy from nutritional rickets were of Black ethnicity. Two children died. Both had a number of known risk factors and presentations to health professionals, but rickets was diagnosed only at postmortem. Dilated cardiomyopathy was sought proactively in one child who presented with a hypocalcaemic seizure and responded well to treatment with calcium and vitamin D.

Biochemical investigations

Due to missing data, differences in biochemical and radiological features of rickets according to age could not be fully described.

Radiology

Radiographs were performed in 107 (85.6%) of 125 included cases, and 79 received expert radiologist review. In 69 cases (87%), radiographs performed at the time of presentation showed clear radiological features of rickets. In four cases, radiographs were performed only after treatment, showing evidence of healing rickets. In three cases, radiographs were insufficient for demonstration of rickets. One case presented near the end of growth did not show signs of rickets but of osteomalacia. Only one included case showed no evidence of rickets but met clinical and biochemical criteria for diagnosis. One case was considered equivocal (changes only convincingly present in the distal ulna).

‘Not meeting the surveillance definition’ cases

Thirty-five cases did not meet the surveillance case definition. In these, nutritional rickets was suspected but not confirmed, or the children had other complications of vitamin D deficiency. Eight cases of nutritional rickets did not meet the criteria for vitamin D deficiency (serum 25-hydroxyvitamin D (25-OHD)<25 nmol/L) but had clinical and radiological features of rickets with raised serum PTH and ALP levels for the age of the child (table 4).

Table 4

Description of ‘not meeting surveillance definition’ cases of nutritional rickets

Management

Colecalciferol was the the most common treatment, with 25 recorded variations in dosage, duration or frequency (and partial or missing data for a further 18 prescriptions). Two children were prescribed alfacalcidol.

A total of 121 patients (96·8%) were being followed up, predominantly by general paediatricians. Only one child had persisting skeletal deformity.

Discussion

Incidence

The estimated annual incidence of nutritional rickets in the UK is 0.48 (95% CI 0.37 to 0.62) cases per 100 000 children under 16 years and is lower than previously thought. It does not confirm the rising rickets cases reported by Goldacre et al,11 the highest hospitalisation rates for rickets over five decades between 2007 and 2011 at 1.78 (95% CI 1.27 to 2.29) children per 100 000 under 15 years, thought to be a reflection of increased incidence or severity of rickets, the changing population, or changes in admission thresholds and diagnostic or coding practices.

The estimated annual incidence of 1.39 (95% CI 1.05 to 1.81) per 100 000 in children under 5 years in the UK also does not support the previous media reports of a rapidly rising number of rickets cases nationally, although the regional variations remain. The West Midlands physician-based survey in 2001 estimated the burden of disease at 7.5 per 100 000 children5 in children under 5 years, and a Scottish study in children aged 2 weeks to 14 years4 showed a rising incidence between 2002 and 2008 with 168 cases of symptomatic vitamin D deficiency identified. Of note, both surveys differ in their reporting of symptomatic vitamin D deficiency, including presentations other than rickets.

Strictly speaking, we have evaluated the incidence of vitamin D deficiency rickets and not nutritional rickets (because we excluded those with 25-OH levels of >25 nmol/L but had raised PTH and radiological features of rickets). We used this strict cut-off of 25-OHD levels of <25 nmol/L in our surveillance definition in order to capture cases of rickets rather than symptomatic vitamin D deficiency, which is a common finding in the population. While there was agreement among UK paediatricians that the risk of vitamin D deficiency rickets increases for 25-OHD levels below 25 nmol/L,12 the exact threshold of 25-OHD below which nutritional rickets occurs is unknown.13 We therefore acknowledge that additional cases of vitamin D insufficiency and radiological rickets as described below would not have been notified by all clinicians.

The annual incidence in the UK is also lower than that in Canada (2.9 per 100 000),14 Australia (4.9 per 100 000),15 New Zealand (2.2 per 100 000)16 and South Denmark (2.9 in 100 000).17 This may be explained by differing case definitions (eg, the Australian study used a much higher vitamin D level of 50 nmol/L and reported symptomatic vitamin D deficiency, including rickets) and international studies conducted prior to embedding of preventative public health guidance. We are aware that our study had a number of limitations, which may have resulted in under-reporting as described further. However, anecdotally, clinicians have reported seeing a lot fewer cases in recent years.

Limitations

Surveillance methodology depends on good report compliance, which the BPSU demonstrates, to maximise case ascertainment. The incidence rate of nutritional rickets presenting to secondary care is a minimal incidence and does not necessarily represent a whole population figure as cases seen in primary care and by non-consultant grade doctors were excluded.

Incidence estimation is not only affected by case underascertainment (in ~20% of reports, no data were received) but by nutritional rickets pathogenesis and management. Using a strict case definition (specifically, 25-OHD of <25 nmol/L) to capture only the osseous complications of nutritional rickets (not symptomatic vitamin D deficiency) may have underestimated those with dietary calcium deficiency and vitamin D insufficiency causing nutritional rickets, as evidenced by the eight additional cases with clear radiographical rickets.

Furthermore, nutritional rickets hypocalcaemic symptoms present at different stages, depending on age,6 chronicity and severity of deficiency, so in practice, rachitic changes may not yet have developed, or clinical and radiological rickets not actively sought.

We also recognise the limitations of single-method reporting. We explored, but did not progress, the linking of the Royal College of General (RCGP) Registry and Hospital Episode Statistics data with BPSU rickets cases due to a number of limitations (patient identifiable data were not captured as part of the RCGP database, costs and timescales).

Clinical Presentation

Significant seasonal variation was not seen in the presentation of nutritional rickets, supporting the recommendation that at-risk children receive vitamin D supplementation throughout the year.18

Black children had a 10-fold, and South Asian children a 5-fold greater incidence of nutritional rickets compared with other ethnic groups under 5 years and the most severe complications (hypocalcaemia, fractures, dilated cardiomyopathy and death). Black and South Asian children have previously been identified as being at high risk of symptomatic vitamin D deficiency4–6 and vitamin D deficiency hypocalcaemic seizures,7 and living in London and West Midlands.9 Worldwide, immigrant and refugee children with darker skin are at higher risk of nutritional rickets2 due to patient, socioeconomic and environmental factors—require more sunlight to make vitamin D; South Asian diets typically contain phytates and oxalates, which impair vitamin D metabolism.19 Genetic susceptibility relating to vitamin D receptor defects has also been suggested20 but not yet fully clarified. Boys were more affected than girls, similar to the vitamin D deficiency hypocalcaemic seizures study.7 In this study, Basetamur and Sutcliffe7 proposed that boys have a greater bone mineral density and rapid growth, with potentially higher demand for calcium and vitamin D, therefore explaining this predilection. This was borne out in our study but has not been evidenced elsewhere.

Children aged 12–23 months were most affected. Unsurprisingly, only six cases were reported in older children, when nutritional rickets usually presents with non-osseous and musculoskeletal symptoms.21

Nutritional rickets resulted in serious sequelae, with missed opportunities to identify them. There were two deaths from dilated cardiomyopathy a rare but well-described complication of vitamin D deficiency.22 The myopathic effects of vitamin D deficiency may not yet be clinically obvious in infants presenting with hypocalcaemic seizures. However, in one child, proactive screening for dilated cardiomyopathy and treatment proved life-saving.

Fracture incidence was similar to the Canadian survey (11%).14 Most children with clinically suspected fractures were mobile but also occurred in non-mobile children as previously reported23 and likely to be an under-represented group. Skeletal surveys were only performed in two cases, so description of the total numbers of fractures occurring in the presence of nutritional rickets was not feasible.

Our data (maternal vitamin D status unknown in most and 77% of children not on supplements) suggest that previous Department of Health (DH) supplementation guidance 24 was not being adhered to, resulting in potentially life-threatening hypocalcaemic seizures in 20 children (10 included cases and 10 cases ‘not meeting surveillance definition’). It complements previous reports of poor adherence to DH guidance relating to pregnant women with high rates of vitamin D deficiency.25

Healthy Start vitamin usage in high-risk populations is low, as previously described.26 Furthermore, some children who were reportedly receiving vitamin D supplements developed rickets, possibly from being born to vitamin D-deficient mothers, or intermittent or non-adherence to supplements. In 2017, the UK had the lowest adherence rates to vitamin D supplements compared with the rest of Europe.27 Both Birmingham28 and Scotland29 previously had targeted and successful public health campaigns to prevent symptomatic vitamin D deficiency, and there is wide support for free universal supplementation for pregnant and lactating women and their children up to the age of 5 years26 to prevent symptomatic vitamin D deficiency and to improve adherence in high-risk groups. Twelve babies presented under the age of 6 months and were not receiving vitamin D supplements, supporting the current recommendation of vitamin D supplementation from birth. While exclusive and prolonged breast feeding is a recognised risk factor for vitamin D deficiency and rickets, one child who was exclusively formula fed developed florid rickets and supported the recommendation that supplementation should occur irrespective of the mode of feeding.30

The huge variation in vitamin D treatment prescriptions, including alfacalcidol (not recommended due to potential toxicity), highlights the need for a simple, standardised guideline for the management of nutritional rickets.

Most children were easily treatable and recovered/recovering during the study period.

Dietary calcium deficiency is likely to have contributed to the causation of nutritional rickets in the eight cases with nutritional rickets and vitamin D insufficiency. Calcium plays a vital role in the pathogenesis of nutritional rickets.30 In our study, nutritional conditions, cow’s milk allergy/intolerance and iron deficiency were commonly associated, and detailed information about dietary calcium was not obtained. This emphasises the importance of ensuring adequate dietary calcium intakes in at-risk populations.

There is no clear cut-off of 25-OHD below which rickets occurs.13 The additional cases of nutritional rickets from calcium deficiency and vitamin D insufficiency support these findings. Had we used 25-OHD <30nmol/L to define Vitamin D deficiency,30 we would still have missed cases of calcium deficiency rickets, reinforcing that both biochemcial and radiological data are important for confirming nutritional rickets cases.30

Conclusion

Nutritional rickets is still presenting in the UK, although the incidence is lower than previously thought. From our data, nutritional rickets could be considered a rare condition. Nevertheless, even one case of nutritional rickets is unacceptable, given that it is an entirely preventable condition. It is resulting in serious complications and unexpected deaths, particularly in Black and South Asian children under 5 years. Both vitamin D and dietary calcium deficiency are playing a role in its pathogenesis. Adherence to current public health prevention policy is failing.

Possible interventions to eradicate nutritional rickets should include (1) detailed clinical histories focussing on both maternal and child risk factors, dietary calcium intake, access and adherence to vitamin D supplements* 27 28 30, proactive screening for antenatal deficiency and correction, (2) education of the public and health professionals to implement preventative advice, and (3) continued successful targeted interventions for families and consideration of culturally acceptable food fortification for high-risk groups.

Future surveillance studies should ensure both radiological and biochemical data to confirm cases of nutritional rickets.

*We support previous recommendations by colleagues27 28 30 to improve vitamin D supplementation, including universal free supplementation for pregnant and lactating women and young children from birth, irrespective of the mode of feeding, together with monitoring of adherence, which could be linked to both routine well child care preventive reviews and opportunistic attendance in primary care and antenatal clinic appointments for pregnant mothers; public health education to improve knowledge and empower purchasing of over-the-counter vitamin D supplements. For the prevention of nutritional rickets, we would reinforce the importance of provision of adequate dietary calcium in addition to vitamin D supplements.

Acknowledgments

We thank the British Paediatric Surveillance Unit for facilitating the data collection and the reporting clinicians, particularly those who reported cases and completed the questionnaires.

References

Footnotes

  • Twitter @rareman4u

  • Contributors PJ: design of the project, review and analysis of data, main author, review of the manuscript. RML: original conception of the project idea; design of the project; collection, recording and statistical analysis of the data; contribution to the write-up of the final manuscript; review of the manuscript. KP: design of the project; collection, recording and statistical analysis of the data; review of the manuscript. ML: collection, recording and statistical analysis of the data; review of the manuscript. AC: design of the project, assessment of cases by an expert panel, review and analysis of the data, contribution to the write-up of the final manuscript, review of the manuscript. ZM: design of the project, assessment of cases by an expert panel, review and analysis of data, review of the manuscript. NS: design of the project, assessment of cases by an expert panel, review of the manuscript. CM: review of the manuscript. HM: assessment of cases by an expert panel, review of the manuscript. MB: original conception of project idea, design of the project, assessment of cases by an expert panel, review of the manuscript, supervision of the main author.

  • Funding An educational grant from the Royal National Orthopaedic Hospital research committee supported by a Vitamin D Nutricia Early Life Nutrition Research grant.

  • Competing interests MB receives funding from the National Institute for Health Research under the Collaborations for Leadership in Applied Health Research and Care programme for North West London. ZM has received honoraria from Internis Pharmaceuticals Ltd.

  • Patient consent for publication Not required.

  • Ethics approval This study was approved by NRES Committee London, West London & GTAC (REC reference: 14/LO/2221, IRAS ID: 144785) and was granted Section 251 HRA-CAG permission (CAG reference: 14/CAG/1042).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement All data relevant to the study are included in the article. As patient data was obtained from reporting clinicians and the National Research Ethics Service approval obtained only for the purposes of the specified activity, the data cannot be made publicly available for other purposes.