Accepted Manuscript

Risk factors of refeeding syndrome in malnourished older hospitalized patients

Maryam Pourhassan, Ingeborg Cuvelier, Ilse Gehrke, Christian Marburger, Mirja

Katrin Modreker, Dorothee Volkert, Hans-Peter Willschrei, Rainer Wirth

PII: S0261-5614(17)30219-4
DOI: 10.1016/[Link].2017.06.008
Reference: YCLNU 3165

To appear in: Clinical Nutrition

Received Date: 7 February 2017

Revised Date: 23 May 2017
Accepted Date: 4 June 2017

Please cite this article as: Pourhassan M, Cuvelier I, Gehrke I, Marburger C, Modreker MK, Volkert D,
Willschrei H-P, Wirth R, Risk factors of refeeding syndrome in malnourished older hospitalized patients,
Clinical Nutrition (2017), doi: 10.1016/[Link].2017.06.008.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
1 Risk factors of refeeding syndrome in malnourished older hospitalized patients

2 Maryam Pourhassan1, Ingeborg Cuvelier2, Ilse Gehrke3, Christian Marburger4, Mirja Katrin

3 Modreker5, Dorothee Volkert6, Hans-Peter Willschrei7 and Rainer Wirth1

1
4 Department of Geriatric Medicine, Marien Hospital Herne, Ruhr-University Bochum, Germany;

PT
2
5 Geriatric Center, Karlstruhe, Germany; 3Department of Internal Medicine IV, Donaueschingen,

6 Germany; 4Christophsbad Clinic, Depatment of Geriatric Rehabilitation, Göppingen, Germany; 5Sana

RI
7 Hanse-Clinic, Department of Geriatric, Wismar, Germany; 6Institute for Biomedicine of Aging,

Friedrich-Alexander-Universität Erlangen-Nürnberg, Nuremberg, Germany; 7Department of Internal

SC
8

9 Medicine, Malteser Hospital St. Josefshospital, Uerdingen, Germany.

10 Address of correspondence:
U
AN
11 Dr. oec. troph. Maryam Pourhassan
M

12 Marien Hospital Herne, Department of Geriatric Medicine

D

13 University Hospital Ruhr-University Bochum, Germany

TE

14 Hölkeskampring 40, D- 44625 Herne, Germany

EP

15 Email: mpourhassan1918@[Link]

16 Financial disclosure: No funding from any organization

C
AC

17 Running title: Risk factors of refeeding syndrome and malnutrition screening

18 Abbreviations: RFS, refeeding syndrome; NRS-2002, Nutritional Risk Screening; MUST, Malnutrition

19 Universal Screening Tool; MNA-SF, Mini Nutritional Assessment-Short Form; WL, weight loss; BMI,

20 body mass index; NICE, National Institute for Health and Clinical Excellence

21 Ms. No. YCLNU-D-17-00117R2

1
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
22 Abstract

23 Background and aims: Despite the high prevalence of malnutrition among older hospitalized

24 persons, it is unknown how many of these malnourished patients are at risk of developing the

25 refeeding syndrome (RFS). In this study, we sought to compare the prevalence and severity of

26 malnutrition among older hospitalized patients with prevalence of known risk factors of RFS.

PT
27 Methods: This cross-sectional multicenter-study investigated older participants who were

RI
28 consecutively admitted to the geriatric acute care ward. Malnutrition screening was conducted using

SC
29 Nutritional Risk Screening (NRS-2002), Malnutrition Universal Screening Tool (MUST) and Mini

30 Nutritional Assessment-Short Form (MNA-SF). The National Institute for Health and Clinical

U
31 Excellence (NICE) criteria were applied for assessing patients at risk of RFS. Weight and height were
AN
32 measured. Degree of weight loss (WL) was obtained by interview. Serum phosphate, magnesium,

33 potassium, sodium, calcium, creatinine and urea were analyzed according to standard procedures.
M

34 Results: The study group comprised 342 participants (222 females) with a mean age of 83.1± 6.8 and
D

35 BMI range of 14.7–43.6 kg/m2. More participants were assessed at risk of malnutrition using NRS-
TE

36 2002 (n=253, 74.0%) compared to MUST (n=170, 49.7%) and MNA-SF (n=191, 55.8%). Of total

37 participants, 239 (69.9%; 157 females) were considered to be at risk of RFS. Based on NRS-2002,
EP

38 75.9% (n=192) of patients at risk of malnutrition are at risk of RFS whereas according to MUST and

39 MNA-SF, 85.9 % (n=146) and 69.1% (n=132) of patients at risk of malnutrition are exposed to high
C

40 risk of RFS, respectively. In addition, the prevalence of risk of RFS is significantly increased with
AC

41 higher score of NRS-2002 and MUST and lower score of MNA-SF. In a stepwise multiple regression

42 analysis, disease severity (38.2%), WL in 3 months (20.3%) and BMI (33.3%) mainly explained

43 variance in NRS-2002, MUST and MNA-SF scores, respectively, in patients with risk of RFS.

2
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
44 Conclusion: Nearly three-quarters of geriatric hospitalized patients with risk of malnutrition

45 demonstrated significant risk of RFS. Therefore, additional screening for risk of RFS in patients

46 screened for malnutrition appears to be abdicable among this population.

47 Keywords: Nutritional screening, malnutrition, older persons, refeeding syndrome

PT
48 Introduction

49 Refeeding syndrome (RFS) can be defined as a clinical complex which encompasses acute fluid and

RI
50 electrolyte disturbances (i.e. phosphate, potassium, magnesium, sodium) and thiamin deficiency

SC
51 associated with metabolic abnormalities in starved individuals as a consequence of reintroduction of

52 feeding whether orally, enterally or parenterally [1]. The symptoms may be various and very

53
U
unspecific and often occur 2-5 days after the start of refeeding [2, 3]. RFS is associated with high
AN
54 mortality; however, it is frequently unrecognized [3, 4]. As there is no precise and unique definition

55 of RFS, unsurprisingly the incidence of RFS is unknown. However, the criteria of the National Institute
M

56 for Health and Clinical Excellence (NICE) guidelines [5] are recognized as a useful tool for identifying
D

57 and screening patients at high risk of developing RFS.

TE

58 RFS may occur usually in malnourished patients undergoing re-feeding after a period of

59 undernutrition [1, 6]. Malnutrition is the most predominant and frequent clinical risk factor among
EP

60 elderly people who are prone for developing RFS due to many underlying comorbidities [4, 7].

61 Results of the previous studies [8, 9] reveal the significant possibility for RFS and it is tempting to
C

62 speculate the link between comorbidity, malnutrition and RFS. Malnutrition is related with high risk
AC

63 of complications such as low quality of life, increased length of hospital stay and higher mortality and

64 morbidity [10, 11]. Although, the prevalence of malnutrition is high among hospitalized elderly

65 people, it remains a widely unrecognized and undertreated problem in this population [12, 13].

66 Accordingly, identification and early treatment of older individuals at risk of malnutrition is essential

67 due to minimizing its occurrence and avoiding mortality and morbidity related with this phenomenon
3
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
68 [14]. Several screening tools have been established to assess the risk and degree of malnutrition for

69 use in elderly /or hospital populations that apply the variables like BMI, recent poor intake and

70 weight loss as well as severity of disease. The Mini-Nutritional Assessment Short Form (MNA-SF) has

71 been proposed as an effective tool for screening the nutritional status of geriatrics across settings

72 [15]. In contrast to MNA-SF, the Malnutrition Universal Screening Tool (MUST) [16] and the

PT
73 Nutritional Risk Screening 2002 (NRS-2002) [17] proposed by the European Society for Clinical

74 Nutrition and Metabolism (ESPEN) for the hospital setting [18, 19] are applicable to all hospital

RI
75 patients, irrespective of age.

SC
76 RFS in not usually considered as a common condition to investigate in older people presenting with

77 undernourishment. Namely, the risk of RFS is mostly not measured among hospitalized or

78
U
institutionalized elderly persons even if malnutrition is evident. Thus, it remains unclear how many of
AN
79 older individuals with malnutrition have to be considered at risk of RFS. As malnutrition is the main
M

80 risk factor for RFS, routine malnutrition screening should also screen for risk of RFS [20].

81 Unfortunately, the criteria of malnutrition screening tools and the NICE criteria for risk of RFS differ
D

82 substantially. In addition, screening for malnutrition and screening for risk of RFS with different tools
TE

83 and criteria appears to be unpractical and somehow redundant. In this study, we sought to compare

84 the prevalence and severity of malnutrition among older hospitalized patients with prevalence of
EP

85 known risk factors of RFS.

C

86 Subjects and Methods

AC

87 Study design and subjects

88 This prospective cross-sectional study was undertaken between July 2015 and February 2016 at six

89 acute geriatric hospital departments in Germany. The study population comprised 342 consecutive

90 hospitalized elderly participants (222 females) with aged between 60 and 100 years. The study

91 protocol had been approved by the ethical committee of Friedrich-Alexander-University, Erlangen-

4
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
92 Nürnberg. Exclusion criteria from the study were: age < 60 years, missing or withdrawn consent of

93 the patients, hypercalcemia, primary hyperparathyroidism and inability to cooperate.

94 Anthropometric measurements

95 Weight was measured in the morning, in light clothing with an empty bladder on a chair scale (Seca

PT
96 Chair Scale 956, Hamburg, Germany).

97 Body weight was assessed in light clothing with an accuracy of 0.1 kg, and height was measured to

RI
98 the nearest 0.5 cm with a stadiometer in first day after admission to hospital. The degree of

SC
99 unintentional weight loss (WL) was obtained either by interviewing the patients and their relatives, if

100 competent, or asking their proxy, where necessary. Calf circumference was measured as part of the

101
U
MNA-SF in a sub-group of 308 patients. Patient’s medication histories were obtained at least within
AN
102 24 hours after hospital admission either through interview or from the medication lists of the general

103 practitioner. All information including malnutrition screening was recorded by attending physician.
M

104 Malnutrition screening tools

D

105 The risk of malnutrition was evaluated using NRS-2002, MUST and MNA-SF on the day after hospital
TE

106 admission.
EP

107 NRS-2002. According to the developing authors [17], the initial screening of NRS-2002 was skipped,

108 because of the high prevalence of malnutrition in the geriatric hospital patients. The final screening
C

109 of the NRS-2002 consists of two criteria: impaired nutritional status based on WL, food intake and
AC

110 BMI (1 - 3 points) and severity of disease (1 - 3 points). Further, for older patients ≥ 70 years, one

111 point to the total score was added. Patients were grouped as no risk (< 3 points) or at risk (≥ 3

112 points).

113 MUST. The MUST [16] includes/considers information on BMI, unintentional WL in the last 3-6

114 months (0 - 2 points), presence of acute disease (0 – 2 points) and absence of food intake > 5 days (0

5
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
115 or 2 points). Participants are stratified as low risk (0 points), medium risk (1 point) and high risk (≥ 2

116 points).

117 MNA-SF. The MNA-SF [15] is a [revised screening tool] reduced version of the Mini-Nutritional

118 Assessment (MNA) which assesses reduction in food intake, WL during last 3 months, mobility,

119 psychological stress and acute diseases, neuropsychological problems and BMI. Patients are classified

PT
120 as having normal nutritional status (12-14 points), at risk of malnutrition (8-11 points) and

RI
121 malnourished (0-7 points).

SC
122 Refeeding syndrome

123 NICE criteria [5] have been applied for assessing patients at risk of RFS. Subjects with at least one of

124
U
the following major parameters (BMI<16 kg/m2, unintended WL >15% in last 3-6 months, no
AN
125 nutrition intake >10 days, low concentration of plasma magnesium, potassium or phosphate before

126 feeding) or with two of the following minor features (BMI<18.5 kg/m2, unintended WL>10 % in last 3-
M

127 6 months, no nutrition intake >5 days, medical history of alcohol or drug abuse) of NICE criteria were
D

128 considered to be at high risk of RFS.

TE

129 Laboratory methods

EP

130 Blood tests were performed on the day of admission at each hospital clinical chemistry laboratory

131 and serum phosphate, magnesium, potassium, sodium, calcium, creatinine and urea were analyzed
C

132 according to local standard procedures. Serum phosphate level < 0.8 mmol/l was defined as
AC

133 hypophosphatemia. In addition, serum magnesium and potassium levels < 0.70 mmol/l and 3.5

134 mmol/l were considered as hypomagnesemia and hypokalemia, respectively.

135 Data analysis

136 The statistical analysis was completed using SPSS statistical software (SPSS Statistics for Windows,

137 IBM Corp, Version 23.0, Armonk, NY, USA). Continuous variables are reported by means and standard

6
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
138 deviations (SDs). Categorical variables have shown as n (%). In order to compare the nutritional

139 screening tools, the results of each tool were classified into two groups: malnourished or at risk of

140 malnutrition (NRS-2002, risk of malnutrition with score ≥ 3; MUST, from medium to high risk of

141 malnutrition with score 1 and more; and MNA-SF, at risk of malnutrition and malnourished with 0-11

142 points) and not at risk of malnutrition (NRS-2002, no risk of malnutrition with score < 3; MUST, low

PT
143 risk of malnutrition with score 0; and MNA-SF, normal nutritional status with scores 12-14).

144 Differences between females and males and between malnourished patients at risk of RFS and not at

RI
145 refeeding risk were analyzed by using an unpaired t test in normally distributed variables. Categorical

SC
146 variables were compared by the Chi square test. Pearson's correlation was applied for normally

147 distributed variables whereas Spearman's correlation was used for nonparametric data. Stepwise

U
148 multiple regression analysis was performed to explain the effect of individual screening tools’
AN
149 questions such as BMI, WL in last 3 months, no food intake, calf circumference and disease severity

150 (as independent variables) on the variance of each screening tools’ scores (as dependent variable). P
M

151 < 0.05 was determined as the limit of significance.

D

152 Results
TE

153 Baseline characteristics and laboratory data of study participants stratified by gender are presented

154 in Table 1. The study group comprised 342 participants (222 females) with a mean age of 83.1± 6.8
EP

155 (age range between 60 and 100 years). According to the respective MNA-SF items, 45.3 % of the
C

156 study participants (155 subjects) have displayed mild degree of dementia, 42 % had no mobility, 43 %
AC

157 were homebound and 15 % could leave their flat. Women were significantly younger than men. The

158 study population showed a wide BMI range from 14.7 to 43.6 kg/m2 with no sex differences

159 (P=0.916). Compared with males, females had significantly lower actual body weight, height, calf

160 circumference and WL during last six months. There were no significant differences in all laboratory

161 data between sexes, except for creatinine with lower values in females than males (P<0.01).

162 According to the guidelines of the NICE criteria, 69.9% of total study population (239 subjects) was
7
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
163 considered to be at risk of RFS. In addition, 51 participants (14.9%) had hypophosphatemia as a

164 hallmark of RFS.

165 The performance of each screening tool to identify malnutrition determined by NRS-2002, MUST and

166 MNA-SF in total population and refeeding risk group are given in Table 2. In total study population,

167 based on NRS-2002, 89 patients (26%) had no risk of malnutrition and 253 subjects (74%) were at risk

PT
168 of malnutrition. Of these 253 malnourished participants, 192 patients (75.9%) are at risk of RFS. The

RI
169 incidence of risk of RFS significantly increased with higher risk score.

SC
170 In addition, the prevalence of patients at nutritional risk according to MUST and MNA-SF tools in

171 total population was as follows: MUST (at medium risk, 17 patients, 5%; at high risk, 153 patients,

U
172 44.7%), MNA-SF (at risk of malnutrition, 153 patients, 44.7%; malnourished, 38 patients, 11.1%;
AN
173 Table 2). According to MUST, of 170 older individuals at medium and high risk of malnutrition, 146

174 patients (85.9%) are at risk of RFS whereas all patients with MUST score 3 or more are exposed to
M

175 high risk of RFS. Moreover, based on MNA-SF, of 191 older subjects at risk or malnourished, 132

176 patients (69.1%) are at risk of RFS while the prevalence of risk of RFS is higher in severe malnourished
D

177 patients with lower score (score <5; Table 2). In addition, the prevalence of hypophosphatemia is
TE

178 excessively raised in malnourished patients ranged from 13.3 % at score of 7 to 44.4% at score of 5.
EP

179 Further, in order to compare the results, participants were stratified into two groups, subjects at risk

180 of malnutrition and subjects not at risk of malnutrition. Figure 1a demonstrates the frequency of
C

181 nutritional status categorization for subjects using MUST, NRS-2002 and MNA-SF. More participants
AC

182 were assessed at risk of malnutrition using NRS-2002 (74%) compared to MUST (49.7%) and MNA-SF

183 (55.8%). In addition, using these nutritional screening tools, the prevalence of risk of RFS is

184 considerably higher compared to not at risk of RFS in malnourished older patients (Figure 1b).

185 There were significant differences in actual body weight, BMI, calf circumference and magnitude of

186 WL (% in last 6 months) between patients at risk and not at risk of malnutrition (all P < 0.001), as

8
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
187 assessed by each nutritional screening tool. Descriptive characteristics and laboratory data of the

188 study participants stratified by three nutritional screening tools for determining patients at high risk

189 of malnutrition and RFS concurrently are given in Table 3. According to each nutritional screening

190 tool, subjects at risk of malnutrition and RFS had significantly higher WL during the last 6 months (P <

191 0.001) and lower magnesium and potassium levels than the malnourished patients not at risk of RFS,

PT
192 with no differences in age and BMI.

RI
193 Associations of each screening tool questions to total scores in malnourished older patients at risk of

194 RFS are shown in Table 4. Significant correlations were observed between each screening tool

SC
195 question to total scores except for mobility and neuropsychological problems. Further, there were no

196 significant differences between MUST scores and disease severity and between MNA-SF scores and

197 no food intake for > 5 days.

U
AN
198 In a stepwise multiple regression analysis on malnourished older patients at risk of RFS, the effects of
M

199 individual screening tools’ questions such as BMI, WL in last 3 months, no food intake, calf

200 circumference and disease severity (as independent variables) on the variance of each screening
D

201 tools’ scores (as dependent variable) were tested (Table 5). According to NRS-2002, disease severity
TE

202 and no food intake explained 38.2% and 24.5% of the variance in NRS-2002 scores, respectively

203 whereas BMI and WL in 3 months explained additional 14%. Based on MUST, WL in 3 months
EP

204 explained 20.3% of the variance in MUST scores and no food intake for > 5 days and BMI explained
C

205 additional 13% and 11.6%, respectively. In MNA-SF, 33.3% of the variance in MNA-SF scores
AC

206 explained mainly by BMI and 8.5% of the variance explained by WL in 3 months and calf

207 circumference (Table 5).

208 Discussion

209 Malnutrition in elderly people is a serious concern and is related with high risk of complications such

210 as increased infection risk, low quality of life and increased mortality and morbidity [10, 11]. Previous

9
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
211 studied have shown that up to 50% of hospitalized older people [21, 22] and 30% of elderly nursing

212 home residents [21] have some degree of malnutrition. In a multi-center prospective study in 13

213 German hospitals, Pirlich et al. reported that of all consecutively admitted patients (mean age 62.2 ±

214 17.4), every fourth patient is malnourished at admission or has a risk of becoming malnourished

215 during hospital stay [23]. In that study 43% of all patients > 70 years and 56% of all patients in the

PT
216 geriatric department were considered to be malnourished.

RI
217 Numerous studies and guidelines have suggested that nutritional screening is fundamental to

218 identify patients at risk of malnutrition or who are malnourished and to manage nutritional problems

SC
219 [24, 25]. Based on current research, malnutrition screening was performed by NRS-2002, MUST and

220 MNA-SF. Our results demonstrated that the prevalence and severity of malnutrition in elderly

221
U
participants varied enormously due to different nutritional screening tools used. However, the
AN
222 overall prevalence of malnutrition is consistently high among our hospitalized patients. In a study of
M

223 geriatric hospitalized patients, Drescher et al demonstrated that 70% of subjects were malnourished

224 or at risk of malnutrition based on MNA and 34% of subjects had moderate to severe risk of
D

225 malnutrition based on NRS-2002 [26]. In other study by Zhou et al, the prevalence of malnutrition in
TE

226 surgical older individuals classified by MNA-SF and NS-2002 were 45% and 38%, respectively [22].

227 Discrepant prevalence of malnutrition or risk of developing undernutrition among different studies
EP

228 may be explained by different kind of nutritional screening tools used and may also partly be due to

229 differences of the study cohorts [23, 27, 28].

C
AC

230 In this study, we found that the prevalence of risk of RFS is significantly increased in older patients at

231 risk of malnutrition as well as in severe malnourished patients based on each tool (Table 2). The

232 significant overlap between the risk factors of malnutrition and the RFS observed in current study as

233 well as previously [9] may explain our findings. Accordingly, it is acceptable to assume the substantial

234 probability of RFS among frail and malnourished older patients [29]. RFS is a potentially fatal

235 condition and is considered to be a serious clinical problem, particularly, in the population of
10
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
236 hospitalized older patients [6]. In the current study, according all three screening tools, three-

237 quarters or more of geriatric hospitalized patients with risk of malnutrition demonstrated significant

238 risk of RFS (see figure 1b). Indeed, if MUST-score is above 1, the risk of RFS is 89.5%. Therefore,

239 additional screening for RFS in patients screened for malnutrition appears to be abdicable among this

240 population. However, it would be worthwhile to mention that based on NICE guidelines [5], older

PT
241 subjects at high risk of RFS should be closely monitored for their vital functions [3], electrolytes (i.e.

242 phosphate, magnesium, potassium, sodium) and thiamine while energy supplementation should

RI
243 start with half of demands, and can then be constantly increased to provide adequate nutrition

SC
244 needs after four to seven days [3].

245 Some limitations of the current study should be discussed. We did not demonstrate the real

246
U
occurrence of RFS and therefore it remains unclear how severe the risk of RFS really was. Moreover,
AN
247 development of electrolyte disturbances over time and any type of nutritional therapy have not been
M

248 addressed in this research. Consequently, more research data are required to provide detailed

249 information about occurrence of RFS in older patients to develop the best preventative strategies.
D

250 Conclusion
TE

251 This study indicates a high prevalence of malnutrition in older patients with different established
EP

252 nutritional screening tools used. Approximately three-quarters of geriatric hospitalized patients with

253 risk of malnutrition demonstrated significant risk of RFS. Accordingly, further screening for risk
C

254 factors of RFS in patients screened for malnutrition appears to be abdicable among this population.
AC

255 Statement of authorship

256 The study was designed by all authors. Data were obtained by RW, IC, IG, CM, MKM and HPW.

257 Statistical analysis was performed by MP. MP, DV and RW prepared the manuscript. All authors read

258 and approved the final manuscript

11
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
259 Conflict of interest

260 The authors declare no conflict of interest.

261 Funding

262 There was no funding source.

PT
263 Acknowledgments

RI
264 With thanks to the working group on nutrition and metabolism of the German Geriatric Society (GGS)

SC
265 for the idea and discussion of the study design.

266 References:

U
AN
267 [1] Solomon SM, Kirby DF. The refeeding syndrome: a review. JPEN Journal of parenteral and enteral
268 nutrition. 1990;14:90-7.
269 [2] Coutaz M. Delirium in malnourished elderly subjects: Could it be refeeding syndrome? European
270 Geriatric Medicine. 2016;7:434-7.
M

271 [3] Crook MA, Hally V, Panteli JV. The importance of the refeeding syndrome. Nutrition (Burbank, Los
272 Angeles County, Calif). 2001;17:632-7.
273 [4] Henderson S, Boyce F, Sumukadas D, Witham MD. Changes in Serum Magnesium and Phosphate
D

274 in Older Hospitalised Patients - Correlation with Muscle Strength and Risk Factors for Refeeding
275 Syndrome. J Nutr Health Aging. 2010;14:872-6.
TE

276 [5] National Institute for Health and Care Excellence: Clinical Guidelines [Internet]. London: National
277 Institute for Health and Care Excellence (UK);. 2003-. Available from:
278 [Link]
279 [6] Rio A, Whelan K, Goff L, Reidlinger DP, Smeeton N. Occurrence of refeeding syndrome in adults
EP

280 started on artificial nutrition support: prospective cohort study. Bmj Open. 2013;3.
281 [7] Slee A, Birch D, Stokoe D. A comparison of the malnutrition screening tools, MUST, MNA and
282 bioelectrical impedance assessment in frail older hospital patients. Clinical Nutrition. 2015;34:296-
C

283 301.
284 [8] Camp MA, Allon M. Severe hypophosphatemia in hospitalized patients. Mineral and electrolyte
AC

285 metabolism. 1990;16:365-8.

286 [9] Kraaijenbrink BVC, Lambers WM, Mathus-Vliegen EMH, Siegert CEH. Incidence of refeeding
287 syndrome in internal medicine patients. Neth J Med. 2016;74:116-21.
288 [10] Norman K, Pichard C, Lochs H, Pirlich M. Prognostic impact of disease-related malnutrition.
289 Clinical Nutrition. 2008;27:5-15.
290 [11] Sorensen J, Kondrup J, Prokopowicz J, Schiesser M, Krähenbühl L, Meier R, et al. EuroOOPS: An
291 international, multicentre study to implement nutritional risk screening and evaluate clinical
292 outcome. Clinical Nutrition. 2008;27:340-9.
293 [12] Volkert D, Saeglitz C, Gueldenzoph H, Sieber CC, Stehle P. Undiagnosed malnutrition and
294 nutrition-related problems in geriatric patients. J Nutr Health Aging. 2010;14:387-92.

12
 Ms. No. YCLNU-D-17-00117R2
ACCEPTED MANUSCRIPT
295 [13] Duran Alert P, Mila Villarroel R, Formiga F, Virgili Casas N, Vilarasau Farre C. Assessing risk
296 screening methods of malnutrition in geriatric patients: Mini Nutritional Assessment (MNA) versus
297 Geriatric Nutritional Risk Index (GNRI). Nutricion hospitalaria. 2012;27:590-8.
298 [14] Tresley J, Sheean PM. Refeeding Syndrome: Recognition Is the Key to Prevention and
299 Management. Journal of the American Dietetic Association. 2008;108:2105-8.
300 [15] Rubenstein LZ, Harker JO, Salva A, Guigoz Y, Vellas B. Screening for undernutrition in geriatric
301 practice: Developing the Short-Form Mini-Nutritional Assessment (MNA-SF). J Gerontol a-Biol.
302 2001;56:M366-M72.
303 [16] Todorovic V, Russell C, Stratton R, Ward J, Elia M. MUST Explanatory Booklet: A Guide to the

PT
304 ‘Malnutrition Universal Screening Tool’(‘MUST’) for Adults:[British Association of Parenteral and
305 Enteral Nutrition’s website]: www. bapen. org. uk/pdfs/must/must_explan. pdf. Assessed in August.
306 2011;29.
307 [17] Kondrup J, Rasmussen HH, Hamberg O, Stanga Z, Ad Hoc EWG. Nutritional risk screening (NRS

RI
308 2002): a new method based on an analysis of controlled clinical trials. Clinical nutrition (Edinburgh,
309 Scotland). 2003;22.
310 [18] Kondrup J, Allison SP, Elia M, Vellas B, Plauth M. ESPEN guidelines for nutrition screening 2002.

SC
311 Clinical nutrition (Edinburgh, Scotland). 2003;22.
312 [19] Kyle UG, Kossovsky MP, Karsegard VL, Pichard C. Comparison of tools for nutritional assessment
313 and screening at hospital admission: A population study. Clinical Nutrition. 2006;25:409-17.
314 [20] Boland K, Solanki D, O’Hanlon C. Prevention and treatment of refeeding syndrome in the acute

U
315 care setting. 2013.
AN
316 [21] Kagansky N, Levy S, Koren-Morag N, Berger D, Knobler H. Hypophosphataemia in old patients is
317 associated with the refeeding syndrome and reduced survival. J Intern Med. 2005;257:461-8.
318 [22] Zhou JD, Wang M, Wang HK, Chi Q. Comparison of two nutrition assessment tools in surgical
319 elderly inpatients in Northern China. Nutrition Journal. 2015;14:68.
M

320 [23] Pirlich M, Schütz T, Norman K, Gastell S, Lübke HJ, Bischoff SC, et al. The German hospital
321 malnutrition study. Clinical Nutrition. 2006;25:563-72.
322 [24] Volkert D, Berner YN, Berry E, Cederholm T, Bertrand PC, Milne A, et al. ESPEN guidelines on
D

323 enteral nutrition: Geriatrics. Clinical Nutrition. 2006;25:330-60.

324 [25] Baek MH, Heo YR. Evaluation of the efficacy of nutritional screening tools to predict malnutrition
TE

325 in the elderly at a geriatric care hospital. Nutrition Research and Practice. 2015;9:637-43.
326 [26] Drescher T, Singler K, Ulrich A, Koller M, Keller U, Christ-Crain M, et al. Comparison of two
327 malnutrition risk screening methods (MNA and NRS 2002) and their association with markers of
328 protein malnutrition in geriatric hospitalized patients. European journal of clinical nutrition.
EP

329 2010;64:887-93.
330 [27] Meijers JMA, Schols JMGA, van der Schueren MAEV, Dassen T, Janssen MAP, Halfens RJG.
331 Malnutrition prevalence in The Netherlands: results of the Annual Dutch National Prevalence
332 Measurement of Care Problems. Brit J Nutr. 2009;101:417-23.
C

333 [28] Vanderwee K, Clays E, Bocquaert I, Gobert M, Folens B, Defloor T. Malnutrition and associated
334 factors in elderly hospital patients: A Belgian cross-sectional, multi-centre study. Clinical Nutrition.
AC

335 2010;29:469-76.
336 [29] Khan LUR, Ahmed J, Khan S, MacFie J. Refeeding Syndrome: A Literature Review. Gastroent Res
337 Pract. 2011;2011.

338

13
 ACCEPTED MANUSCRIPT
Table 1. Characteristics of the study participants stratified by gender (Mean ± SD)

All Females Males

(n=342) (n=222; 65 %) (n=120; 35 %)
Age (y) 83.1 ± 6.8 83.9 ± 6.5 81.4 ± 7.0**
Height (m) 1.6 ± 0.1 1.6 ± 0.1 1.7 ± 0.1***
Actual body weight (kg) 72.3 ± 16.9 68.5 ± 16.5 79.1 ± 15.5***
BMI (kg/m2) 26.3 ± 5.4 26.3 ± 5.7 26.3 ± 4.7

PT
Calf circumference 33.2 ± 4.4 32.7 ± 4.3 34.0 ± 4.5*
WL in 1 months (kg) 0.8 ± 1.5 0.9 ± 1.1 1.2 ± 2.0*
WL in 2 months (kg) 1.0 ± 1.8 0.8 ± 1.5 1.4 ± 2.3*
WL in 3 months (kg) 1.2 ± 2.1 0.9 ± 1.7 1.7 ± 2.7**

RI
WL in 6 months (kg) 1.5 ± 2.6 1.0 ± 1.9 2.3 ± 3.6***
Laboratory data (mmol/l)
Phosphate 1.0 ± 0.2 1.0 ± 0.3 1.0 ± 0.2

SC
Magnesium 0.8 ± 0.1 0.7 ± 0.2 0.8 ± 0.1
Potassium 4.1 ± 0.6 4.1 ± 0.6 4.2 ± 0.6
Sodium 138.6 ± 8.6 138.1 ± 10.2 139.5 ± 4.4

U
Calcium 2.3 ± 0.2 2.3 ± 0.2 2.2 ± 0.1
Creatinine 1.2 ± 0.6 1.1 ± 0.6 1.3 ± 0.7**
AN
Urea (mg/dl) 53.2 ± 32.8 50.7 ± 31.3 58.0 ± 35.2
Risk of RFS (n, %) (239, 69.9 %) (157, 65.5 %) (82, 34.5 %)
M

WL; weight loss, Risk of RFS; risk of refeeding syndrome according to the guidelines of NICE criteria. Calf
D

circumference was measured in sub-group of 308 older patients (201 females).

TE

*P < 0.05, **P < 0.001, ***P < 0.001 difference between gender (unpaired t test)
C EP
AC
 ACCEPTED MANUSCRIPT
Table 2. Nutritional status categorization (n; %) in total study population (n=342) and according to

the NICE criteria

NICE†
Total population At risk of RFS Not at risk of RFS
Malnutrition screening tools
(n=342; F=222) (n=239; F=156) (n=103; F=66)
NRS-2002 (n; %)

PT
No risk of malnutrition (score < 3)
Score 2 89; 26.0 47; 52.8 42; 47.2
At risk of malnutrition (score ≥ 3) 253; 74.0 192; 75.9 61; 24.1

RI
Score 3 101; 29.5 69; 68.3 32; 31.7
Score 4 74; 21.6 57; 77.0 17; 23.0
Score 5 56; 16.4 45; 80.4 11; 19.6

SC
Score 6 20; 5.8 19; 95.0 1; 5.0
Score 7 2; 0.6 2; 100 -
MUST (n; %)
Low risk (score 0) 172; 50.3 93; 54.1 79; 45.9

U
Medium risk (score 1) 17; 5.0 9; 52,9 8; 47,1
High risk (score 2 or more) 153; 44.7 137, 89.5 16; 10.5
AN
Score 2 113; 33.0 97; 85.8 16; 14.2
Score 3 24; 7.0 24, 100 -
Score 4 12; 3.5 12; 100 -
Score 5 3; 0.9 3; 100 -
M

Score 6 1; 0.3 1; 100 -

MNA-SF (n; %)
Normal nutritional status (Score 12-14) 151; 44.2 107; 70.9 44; 29.1
D

Score 14 31; 9.1 20; 64.5 11; 35.5

Score 13 61; 17.8 44; 72.1 17; 27.9
TE

Score 12 49; 14.3 34; 69.4 15; 30.6

At risk of malnutrition (Score 8-11) 153; 44.7 104; 68.0 49; 32.0
Score 11 52; 15.2 34; 65.4 18; 34.6
EP

Score 10 39; 11.4 26; 66.7 13; 33.3

Score 9 31; 9.1 20; 64.5 11; 35.5
Score 8 31; 9.1 24; 77.4 7; 22.6
Malnourished (Score 0-7) 38; 11.1 28, 73.7 10; 26.3
C

Score 7 21; 6.1 15; 71.4 6; 28.6

Score 6 9; 2.6 6; 66.7 3; 33.3
AC

Score 5 5; 1.5 4; 80 1; 20
Score 4 2; 0.6 2; 100 -
Score 3 1; 0.3 1; 100 -

†
According to the guidelines of NICE criteria, F; females.
 ACCEPTED MANUSCRIPT

Table 3. Characteristics and laboratory data of the study population (n=342) stratified by three different nutritional screening tools (Mean ± SD)

NRS-2002 MUST MNA-SF

PT
At Risk of At risk of At Risk of At risk of At Risk of At risk of
malnutrition & RFS malnutrition & no RFS malnutrition & RFS malnutrition & no malnutrition & RFS malnutrition & no RFS
(75.9%) (24.1%) (85.9%) RFS (14.1%) (69.1%) (30.9%)

RI
(n=192; F=122) (n=61; F=42) (n=146; F=95) (n=24; F=18) (n=132; F=82) (n=59; F=40)
Age (y) 83.1 ± 6.9 83.6 ± 6.7 83.7 ± 7.2 83.2 ± 7.0 83.8 ± 7.3 84.8 ± 6.9

SC
Height (m) 1.6 ± 0.1 1.6 ± 0.1 1.6 ± 0.1 1.6 ± 0.1 1.6 ± 0.1 1.6 ± 0.1
Actual body weight (kg) 70.9 ± 18.0 68.4 ± 14.2 68.9 ± 17.4 66.0 ± 13.7 66.0 ± 16.9 67.2 ± 14.0
2
BMI (kg/m ) 25.9 ± 5.6 25.3 ± 4.6 25.3 ± 5.6 24.5 ± 4.1 23.9 ± 1.1 24.8 ± 4.0

U
Calf circumference 32.4 ± 4.7 32.9 ± 3.7 32.2 ± 4.7 31.3± 3.6 31.0 ± 4.6 31.5 ± 3.2
WL in 1 months (kg) 1.2 ± 1.8 0.6 ± 1.1** 1.4 ± 2.0 1.1 ± 1.4 1.3 ± 1.7 0.6 ± 1.1***

AN
WL in 2 months (kg) 1.5 ± 2.2 0.7 ± 1.2** 1.8 ± 2.4 1.4 ± 1.5 1.7 ± 2.2 0.7 ± 1.2***
WL in 3 months (kg) 1.7 ± 2.6 0.7 ± 1.2*** 2.1 ± 2.8 1.4 ± 1.6* 2.0 ± 2.6 0.7 ± 1.3***
WL in 6 months (kg) 2.2 ± 3.2 0.8 ± 1.3*** 2.6 ± 3.5 1.6 ± 1.7* 2.4 ± 3.4 0.8 ± 1.4***

M
Laboratory data (mmol/l)
Phosphate 1.0 ± 0.3 1.0 ± 0.2 1.0 ± 0.3 1.0 ± 0.1 1.0 ± 0.3 1.1 ± 0.2*
Magnesium 0.8 ± 0.1 0.8 ± 0.1*** 0.8 ± 0.1 0.8 ± 0.1* 0.8 ± 0.1 0.8 ± 0.1***

D
Potassium 4.0 ± 0.6 4.3 ± 0.4*** 4.0 ± 0.6 4.3 ± 0.5* 4.1 ± 0.6 4.3 ± 0.5**
Sodium 138.9 ± 5.8 139.8 ± 3.9 138.8 ± 6.3 139.3 ± 3.9 138.8 ± 6.0 139.1 ± 4.2

TE
Calcium 2.2 ± 0.2 2.3 ± 0.1 2.2 ± 0.2 2.3 ± 0.1 2.2 ± 0.2 2.3 ± 0.1
Creatinine 1.3 ± 0.7 1.2 ± 0.8 1.2 ± 0.7 1.0 ± 0.6 1.2 ± 0.7 1.1 ± 0.5
Urea (mg/dl) 57.7 ± 38.2 49.0 ± 27.7 55.5 ± 33.4 41.2 ± 18.9** 52.2 ± 29.1 49.8 ± 27.0
EP
F; females, NRS-2002 (no risk of malnutrition with score < 3 and at risk of malnutrition with score ≥ 3), MUST (low risk of malnutrition with score 0 and from
C
AC

medium to high risk of malnutrition with score 1 and more), MNA-SF (normal nutritional status with 12-14 points; at risk of malnutrition and malnourished with 0-

11 points), RFS; refeeding syndrome, WL; weight loss. Calf circumference was measured in sub-group of 308 older patients (201 females).

*P < 0.05; **P < 0.01; *** P <0.001 difference between groups in each nutritional screening tool (unpaired t test).
 ACCEPTED MANUSCRIPT
Table 4. Correlations of each malnutrition screening tool questions to total malnutrition screening scores in

malnourished older patients at risk of refeeding syndrome

NRS-2002 (n=192) MUST (n=146) MNA-SF (n=132)

r P value r P value r P value
2
BMI (kg/m ) -0.46 <0.001 -0.31 <0.001 0.61 <0.001
WL in 3 months 0.29 <0.001 0.30 <0.001 -0.23 0.007

PT
WL in 6 months 0.27 <0.001 0.40 <0.001 -0.22 0.010
Calf circumference -0.37 <0.001 -0.26 0.002 0.55 <0.001
mobility -0.11 0.108 -0.170 0.210 0.01 0.839
Disease severity 0.52 <0.001 0.18 0.028 -0.02 0.777

RI
Neuropsychological problems -0.09 0.202 -0.02 0.984 0.07 0.417
No food intake in last week 0.57 <0.001 0.27 0.001 -0.18 0.034
no food intake for > 5 days 0.17 0.014 0.26 0.001 -0.08 0.341

SC
NRS-2002 (no risk of malnutrition with score < 3 and at risk of malnutrition with score ≥ 3), MUST (low risk of

U
malnutrition with score 0 and medium to high risk of malnutrition with score 1 and more), MNA-SF (normal
AN
nutritional status with 12-14 points; at risk of malnutrition and malnourished with 0-11 points), WL; weight

loss.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 5. Results of multiple regression analysis with malnutrition screening tools’ scores as dependent

variable and malnutrition screening tools ‘questions as independent variables in patients at risk of refeeding

syndrome (n=239)

Models NRS-2002 (n=192)

Variables R R2 SEE P

PT
Disease severity 0.618 0.378 0.828 <0.001
No food intake in last week 0.792 0.623 0.645 <0.001
BMI (kg/m2) 0.864 0.742 0.533 <0.001

RI
WL during the last 3 months 0.876 0.761 0.513 <0.001
Variables MUST (n=146)
BMI (kg/m2) 0.341 0.116 0.808 <0.001

SC
No food intake for > 5days 0.496 0.246 0.750 <0.001
WL during the last 3 months 0.670 0.448 0.643 <0.001
Variables MNA-SF (n=132)
BMI (kg/m2)

U
0.577 0.333 1.514 <0.001
WL during the last 3 months 0.605 0.366 1.483 0.018
AN
Calf circumference 0.646 0.418 1.427 0.002

NRS-2002 (no risk of malnutrition with score < 3 and at risk of malnutrition with score ≥ 3), MUST (low risk of
M

malnutrition with score 0 and medium to high risk of malnutrition with score 1 and more), MNA-SF (normal
D

nutritional status with 12-14 points; at risk of malnutrition and malnourished with 0-11 points), WL; weight
TE

loss, SEE; Standard Error of the Estimate.

C EP
AC
 ACCEPTED MANUSCRIPT

a
At risk of malnutrition
b At risk of RFS
Not at risk of malnutrition Not at risk of RFS

PT
100 100
***
90 90 ***
***
Percentage of patients

RI
80 80

Percentage of patients
70 70

SC
60 60
50 50
40 40

U
30 30

AN
20 20
10 10

M
0 0
NRS-2002 MUST MNA-SF NRS-2002 MUST MNA-SF

D
Nutritional screening tools Nutritional screening tools

TE
EP
Figure 1. a) Prevalence of risk of malnutrition according to NRS-2002, MUST and MNA-SF in total study population (n=342) and
C

b) Prevalence of risk of refeeeding syndrom (RFS) in malnourished older patients using NRS-2002 (n=192 at risk of RFS, n=61 not
AC

at risk of RFS), MUST (n=146 at risk of RFS, n=24 not at risk of RFS) and MNA-SF (n=132 at risk of RFS, n=59 not at risk of RFS).

***P < 0.001 difference between at risk of RFS and Not at risk RFS within groups (unpaired t test).