SIR—There is increasing interest in the influence of vitamin D on tissues other than bone [ 1 ]. Active vitamin D (1,25-dihydroxyvitamin D 3 ) is an important regulator of the immune system and has been most studied with regard to autoimmunity, where it acts as an immunosuppressant [ 2 ]. The influence of 1,25-dihydroxyvitamin D 3 on infection in vitro and in animal models has been little studied [ 3 ]. Mouse studies have suggested that 1,25-dihydroxyvitamin D 3 enhances Mycobacterium bovis killing [ 4 ], but increases the susceptibility to Toxoplasma gondii [ 5 ]. 1,25-dihyroxyvitamin D 3 directly regulates antimicrobial peptide gene expression, which could enhance host defence against infection [ 6–8 ].

Case control studies have found associations between vitamin D deficiency and tuberculosis in Asians in the UK [ 9 ] and severe acute lower respiratory and gastrointestinal infections in children in Ethiopia, Turkey, India and Jordan [ 10–13 ]. Peripheral blood T lymphocytes were significantly decreased in Turkish children with rickets [ 14 ], and children with rickets were found to have reduced neutrophil motility [ 15 ] and phagocytosis [ 16 ].

We have been unable to find any randomised controlled trials of vitamin D supplementation, which have examined the incidence of infection in at-risk groups. Rehman [ 17 ] reported reductions in respiratory infections and febrile illnesses in a group of Indian children with high alkaline phosphatase levels, but no clinical evidence of rickets as a result of vitamin D and calcium supplementation. A randomised trial of co-administration of intramuscular 1,25-dihydroxyvitamin D 3 with influenza vaccine was not found to enhance humoral immunity in healthy young volunteers, whose vitamin D status was not described [ 18 ].

At a conservative estimate, at least 8% of UK community-dwelling older people, and 32% of older people living in institutions have vitamin D deficiency (serum 25-hydroxyvitamin D 3 levels below 25 nmol/l) [ 19 , 20 ]. However, there is some evidence that the optimal serum 25-hydroxyvitamin D 3 level for health could be as high as 90–100 nmol/l [ 21 ].

As an adjunct to the RECORD trial [ 22 ], a blinded, randomised, placebo-controlled trial of oral vitamin D 3 and/or calcium supplementation for the secondary prevention of osteoporotic fractures, we examined whether vitamin D was associated with a reduction in self-reported infections and antibiotic use.

Methods

Five thousand two hundred and ninety-two participants were randomised within a factorial design to 800 IU (20 µg) daily vitamin D 3 , 1,000 mg calcium (calcium carbonate), both, or placebo, and followed up for 24–62 months. The trial was based in 21 centres in England and Scotland. Ethical approval was obtained from the Multicentre Research Ethics Committee for Scotland and each centre's Local Research Ethics Committee. Participants gave written informed consent. Full details and main results of the trial are reported elsewhere [ 22 ].

In March 2002, when 25-hydroxyvitamin D 3 levels are lowest in older people in the northern hemisphere [ 19 ], all trial participants who were alive, or had not withdrawn from filling in questionnaires were sent a reply-paid postal questionnaire. This asked if they had had an infection or received antibiotics in the previous week.

We compared all participants who had been randomised to take vitamin D 3 with all participants who had not been randomised to D 3 (i.e. intention-to-treat analysis), using multiple logistic regression (adjusted for the trial minimisation factors of gender, age, type of enrolling fracture and time since fracture).

Results

Three thousand four hundred and forty-four participants responded to the questionnaire at a median (interquartile range) time of 18 (11–25) months since randomisation. Based on questionnaire responses, at least 55% of trial participants were still taking their tablets. Of the respondents randomised to vitamin D 3 , 17.2% (300/1,740) reported an infection, compared with 18.8% (321/1,704) on placebo (adjusted odds ratio 0.90, 95% confidence interval 0.76 to 1.07, P  = 0.23) (Table 1 ). A percentage of 6.4 (111/1,737) randomised to vitamin D 3 reported antibiotic use compared with 7.5% (128/1,703) on placebo (adjusted odds ratio 0.84, 95% CI 0.64 to 1.09, P  = 0.18).

Table 1

Baseline characteristics and reported infections and antibiotic use during window week

  With vitamin D 3 ( n  = 2, 649)   Without vitamin D 3 ( n  = 2, 643)  Odds ratio(95% CI) P value  
Participant characteristics at recruitment     
Age—mean [SD]  77 [ 6 ]   77 [ 6 ]    
Sex – n (%) female  2,240 (84.6) 2,241 (84.8)   
Type of enrolling fracture – n (%)      
proximal femur 459 (17.3) 445 (16.8)   
distal forearm 924 (34.9) 922 (34.9)   
other fractures 1,266 (47.8) 1,276 (48.3)   
Weight—mean [SD] (kg)  65 [ 13 ]   65 [ 12 ]    
Current smoker – n (%)  298 (11.3) 320 (12.1)   
Daily physical activity—could walk out-doors unaccompanied – n (%)  2,492 (94.4) 2,487 (94.3)   
Diabetic n (%)  222 (8.4) 202 (7.6)   
Current use of oral steroids ⩾7.5 mg prednisolone daily 49 (1.9) 44 (1.7)   
Sent questionnaire for window week  ( n  = 1, 966)   ( n  = 1, 955)    
responded to infection question  ( n  = 1, 740)   ( n  = 1, 704)    
reported infection − n (%)  300 (17.2) 321 (18.8) 0.90 (0.76–1.07) 0.23 
responded to antibiotic question  ( n  = 1,737)   ( n  = 1, 703)    
reported antibiotic use − n (%)  111 (6.4) 128 (7.5) 0.84 (0.64–1.09) 0.18 
Per protocol analysis     
responded to infection question  ( n  = 1, 090)   ( n  = 1, 036)    
reported infection − n (%)  165 (15.1) 190 (18.3) 0.80 (0.64–1.01) 0.06 
responded to antibiotic question  ( n  = 1, 086)   ( n  = 1, 040)    
reported antibiotic use − n (%)  58 (5.3) 73 (7.0) 0.74 (0.52–1.06) 0.10 
  With vitamin D 3 ( n  = 2, 649)   Without vitamin D 3 ( n  = 2, 643)  Odds ratio(95% CI) P value  
Participant characteristics at recruitment     
Age—mean [SD]  77 [ 6 ]   77 [ 6 ]    
Sex – n (%) female  2,240 (84.6) 2,241 (84.8)   
Type of enrolling fracture – n (%)      
proximal femur 459 (17.3) 445 (16.8)   
distal forearm 924 (34.9) 922 (34.9)   
other fractures 1,266 (47.8) 1,276 (48.3)   
Weight—mean [SD] (kg)  65 [ 13 ]   65 [ 12 ]    
Current smoker – n (%)  298 (11.3) 320 (12.1)   
Daily physical activity—could walk out-doors unaccompanied – n (%)  2,492 (94.4) 2,487 (94.3)   
Diabetic n (%)  222 (8.4) 202 (7.6)   
Current use of oral steroids ⩾7.5 mg prednisolone daily 49 (1.9) 44 (1.7)   
Sent questionnaire for window week  ( n  = 1, 966)   ( n  = 1, 955)    
responded to infection question  ( n  = 1, 740)   ( n  = 1, 704)    
reported infection − n (%)  300 (17.2) 321 (18.8) 0.90 (0.76–1.07) 0.23 
responded to antibiotic question  ( n  = 1,737)   ( n  = 1, 703)    
reported antibiotic use − n (%)  111 (6.4) 128 (7.5) 0.84 (0.64–1.09) 0.18 
Per protocol analysis     
responded to infection question  ( n  = 1, 090)   ( n  = 1, 036)    
reported infection − n (%)  165 (15.1) 190 (18.3) 0.80 (0.64–1.01) 0.06 
responded to antibiotic question  ( n  = 1, 086)   ( n  = 1, 040)    
reported antibiotic use − n (%)  58 (5.3) 73 (7.0) 0.74 (0.52–1.06) 0.10 

‘Per protocol’ analyses, based on tablet taking when completing the questionnaire, showed smaller odds ratios for infection (0.80, 95% CI 0.64 to 1.01, P  = 0.06) and antibiotic use (0.74, 95% CI 0.52 to 1.06, P  = 0.10).

Adverse events such as hypercalcaemia, renal stones and renal impairment were rare and did not differ between those people receiving vitamin D 3 or not.

Discussion

Although the observed differences were consistent with vitamin D reducing the risk of infection, the results were not statistically significant. ‘Per protocol’ analyses showed smaller odds ratios for infection and antibiotic use, but the concern about such ‘per protocol’ analyses is that they are prone to bias, however, rates of tablet taking were similar for vitamin D 3 and placebo groups. Non-compliance may therefore have reduced the protective effect. Our questions may also have been too crude, and it might have been preferable to ask participants to keep diaries for longer and to record the length, type and severity of infections.

Only 6% of the RECORD trial participants could not walk out-doors unaccompanied: hence, our participants were likely to have had higher sunlight exposure for vitamin D manufacture in the skin than less mobile older people and be less likely to benefit. The 25-hydroxyvitamin D 3 level from a small sample of 60 trial participants in Southampton and Newcastle, measured from February to July before supplementation averaged 38 (SD 16) nmol/l by high-performance liquid chromatography [ 22 ], similar to the mean level for older people in institutions in recent UK surveys [ 19 , 20 ]. After 1 year of supplementation in these participants 25-hydroxyvitamin D 3 was 62 (SD 16) nmol/l.

Based on data from studies of bone mineral density, lower-extremity function, falls, fractures, colorectal cancer and dental health, Bischoff-Ferrari and colleagues have argued that serum concentrations of 25-hydroxyvitamin D 3 should be 75 nmol/l or more, with optimal levels at least 90–100 nmol/l [ 21 ]. In older people, an intake of ⩾1000 IU (25 µg) vitamin D 3 could bring 25-hydroxyvitamin D 3 to 75 nmol/l in more than 50% of the population [ 21 ]. However, the studies contributing these data used different methods for determining serum 25-hydroxyvitamin D 3 and rarely reported compliance. Currently available assays for 25-hydroxyvitamin D 3 do not agree well, so that assay-specific decision limits have been suggested for assessing compliance and 25-hydroxyvitamin D 3 status [ 23 ]. Further research is needed to assess the effect of a higher dose of vitamin D on infections in populations at high risk of insufficiency.

Key points

  • 1,25-dihyroxyvitaminD 3 is an important regulator of the immune system.

  • Vitamin D deficiency is common in older people in the UK, particularly in people living in institutions.

  • Daily supplementation with 800 IU vitamin D 3 (20 μg) was well tolerated by older people living in the community, with no detectable adverse effects.

  • Supplementation was associated with fewer self-reported infections and antibiotics, but the estimated 10–15% reduction was not statistically significant.

  • Further randomised trials of vitamin D and infections are warranted in at-risk populations.

Conflict of interest statement

None declared for the authors. Conflict of interest statements for other members of the RECORD Trial Group are given in reference number 22 .

Record Trial Group

Record Trial Management Group

Health Services Research Unit, University of Aberdeen, Aberdeen, UK (AM Grant, A Avenell, MK Campbell, AM McDonald, GS MacLennan, GC McPherson); University of Southampton, UK (FH Anderson); MRC Epidemiology Resource Centre, University of Southampton, UK (C Cooper); Freeman Hospital, Newcastle-Upon-Tyne (RM Francis); Centre for Health Services Research and Business School—Economics, University of Newcastle-Upon-Tyne, UK (C Donaldson); The Hull York Medical School, Hull, UK (WJ Gillespie); Royal Infirmary of Edinburgh, UK (CM Robinson); Department of Health Sciences, York, UK (DJ Torgerson); Queens Medical Centre, Nottingham, UK (WA Wallace).

Record Trial Biochemical Analyses

WD Fraser, Royal Liverpool University Hospital, Liverpool.

Further members of the RECORD Trial Group are listed in reference22

We thank the patients who took part in the RECORD study, without whose help this study would not have been possible. The MRC funded the central organisation of RECORD, and Shire Pharmaceuticals funded the drugs, which were co-funded and manufactured by Nycomed. Shire Pharmaceuticals and Nycomed were given the opportunity to comment on the penultimate version of this trial report. However, the trial was conducted, analysed and reported independently of all the funding parties.

References

1.
Zittermann
A
Vitamin D in preventive medicine: are we ignoring the evidence?
Br J Nutr
 , 
2003
, vol. 
89
 (pg. 
552
-
72
)
2.
Hayes
CE
Nashold
FE
Spach
KM
Pedersen
LB
The immunological functions of the vitamin D endocrine system
Cell Mol Biol
 , 
2003
, vol. 
49
 (pg. 
277
-
300
)
3.
Cantorna
MT
Zhu
Y
Froicu
M
Wittke
A
Vitamin D status, 1,25-dihydroxyvitamin D 3 , and the immune system
Am J Clin Nutr
 , 
2004
, vol. 
80
 (pg. 
1717S
-
20
)
4.
Waters
WR
Palmer
MV
Nonnecke
BJ
Whipple
DL
Horst
RL
Mycobacterium bovis infection of vitamin D-deficient NOS2 −/− mice
Microb Pathog
 , 
2004
, vol. 
36
 (pg. 
11
-
7
)
5.
Rajapakse
R
Mousli
M
Pfaff
AW
, et al.  . 
1,25-dihydroxyvitamin D3 induces splenocyte apoptosis and enhances BALB/c mice sensitivity to toxoplasmosis
J Steroid Biochem Mol Biol
 , 
2005
, vol. 
96
 (pg. 
179
-
85
)
6.
Wang
T-T
Nestel
FP
Bourdeau
V
, et al.  . 
Cutting edge: 1,25-dihydroxyvitamin D 3 is a direct inducer of antimicrobial peptide gene expression
J Immunol
 , 
2004
, vol. 
173
 (pg. 
2909
-
12
)
7.
Gombart
AF
Borregaard
N
Koeffler
HP
Human cathelicidin antimicrobial peptide CAMP gene is a direct target of the vitamin D receptor and is strongly up-regulated in myeloid cells by 1,25-dihydroxyvitamin D 3
FASEB J
 , 
2005
, vol. 
19
 (pg. 
1067
-
77
)
8.
Liu
PT
Stenger
S
Li
H
, et al.  . 
Toll-like receptor triggering of a vitamin-D mediated human antimicrobial response
Science
 , 
2006
, vol. 
311
 (pg. 
1770
-
3
)
9.
Wilkinson
RJ
Llewelyn
M
Toossi
Z
, et al.  . 
Influence of vitamin D deficiency and vitamin D receptor polymorphisms on tuberculosis among Gujarati Asians in west London: a case-control study
Lancet
 , 
2000
, vol. 
355
 (pg. 
618
-
21
)
10.
Beser
E
Cakmakci
T
Factors affecting the morbidity of vitamin D deficiency rickets and primary protection
East Afr Med J
 , 
1994
, vol. 
71
 (pg. 
358
-
62
)
11.
Muhe
L
Lulseged
S
Mason
KE
Simoes
EA
Case-control study of the role of nutritional rickets in the risk of developing pneumonia in Ethiopian children
Lancet
 , 
1997
, vol. 
349
 (pg. 
1801
-
4
)
12.
Wayse
V
Yousafzai
A
Mogale
K
Filteau
S
Association of subclinical vitamin D deficiency with severe acute lower respiratory tract infection in Indian children under 5y
Eur J Clin Nutr
 , 
2004
, vol. 
58
 (pg. 
563
-
7
)
13.
Najada
AS
Habashneh
MS
Khader
M
The frequency of nutritional rickets among hospitalized infants and its relation to respiratory diseases
J Trop Paed
 , 
2004
, vol. 
50
 (pg. 
364
-
8
)
14.
Yener
E
Coker
C
Cura
A
Keskinoglu
A
Mir
S
Lymphocyte subpopulations in children with vitamin D deficient rickets
Acta Paed Japon
 , 
1995
, vol. 
37
 (pg. 
500
-
2
)
15.
Lorente
F
Fontan
G
Jara
P
Casas
C
Garcia-Rodriguez
MC
Ojeda
JA
Defective neutrophil motility in hypovitaminosis D rickets
Acta Paediatr Scand
 , 
1976
, vol. 
65
 (pg. 
695
-
9
)
16.
Stroder
J
Kasal
P
Evaluation of phagocytosis in rickets
Acta Paediatr Scand
 , 
1970
, vol. 
59
 (pg. 
288
-
92
)
17.
Rehman
PK
Sub-clinical rickets and recurrent infection
J Trop Paed
 , 
1994
, vol. 
40
 pg. 
58
 
18.
Kriesel
JD
Spruance
J
Calcitriol 1,25-dihydroxy-vitamin D3 coadministered with influenza vaccine does not enhance humoral immunity in human volunteers
Vaccine
 , 
1999
, vol. 
17
 (pg. 
1883
-
8
)
19.
Finch
S
Doyle
W
Lowe
C
, et al.  . 
National Diet and Nutrition Survey: People aged 65 Years and Over
1998
London
The Stationery Office
 
Volume 1: Report of the Diet and Nutrition Survey
20.
Hirani
V
Primatesta
P
Vitamin D concentrations among people aged 65 years and over living in private households and institutions in England: population survey
Age Ageing
 , 
2005
, vol. 
34
 (pg. 
485
-
91
)
21.
Bischoff-Ferrari
HA
Giovannucci
E
Willett
WC
Dietrich
T
Dawson-Hughes
B
Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes
Am J Clin Nutr
 , 
2006
, vol. 
84
 (pg. 
18
-
28
)
22.
The RECORD Trial Group
Oral vitamin D3 and calcium for secondary prevention of low-trauma fractures in elderly people Randomised Evaluation of Calcium and/OR vitamin D, RECORD: a randomised placebo-controlled trial
Lancet
 , 
2005
, vol. 
365
 (pg. 
1621
-
8
)
23.
Glendenning
P
Taranto
M
Noble
JM
, et al.  . 
Current assays overestimate 25-hydroxyvitamin D 3 and underestimate 25-hydroxyvitamin D 2 compared with HPLC: need for assay-specific decision limits and metabolite-specific assays
Ann Clin Biochem
 , 
2006
, vol. 
43
 (pg. 
23
-
30
)

Comments

0 Comments