Impaired Nutritional Condition After Stroke From the Hyperacute to the Chronic Phase: A Systematic Review and Meta-Analysis

Abstract

Background: Malnutrition is common after stroke and can affect rehabilitation and healthcare costs. A comprehensive overview of stroke patients' nutritional condition from the hyperacute to the chronic phase is lacking. This systematic review aimed to investigate the prevalence of impaired nutritional condition (INC) across the continuum of care in specific phases after stroke.

Methods: CAB ABSTRACTS, Embase, MEDLINE, were used to collect studies published between 01-01-1999 and 26-08-2020. Primary and secondary outcomes were prevalence of INC and prevalence of malnutrition, respectively. Exploratory outcomes were prevalence of INC at follow-up, nutritional examination methods, prevalence of dysphagia, stroke severity, adverse events, and continent-specific prevalence of INC. A random-effects meta-analysis model was used to estimate the phase-specific pooled prevalence of INC and malnutrition.

Results: The dataset consisted of 78 study groups selected over a total of 1,244 identified records. The pooled prevalence of INC and malnutrition were 19% (95%CI:7-31) (N = 4) and 19% (95%CI:9-29) (N = 3), 34% (95%CI:25-43) (N = 34) and 26% (95%CI:18-35) (N = 29), 52% (95%CI:43-61) (N = 34) and 37% (95%CI:28-45) (N = 31), 21% (95%CI:12-31) (N = 3) and 11% (95%CI:0-24) (N = 3) and 72% (95%CI:41-100) (N = 3) and 30% (95%CI:0-76) (N = 2) in the hyperacute, acute, early subacute, late subacute, and chronic phase, respectively.

Conclusion: INC and malnutrition are highly prevalent in all stages of stroke care. Since malnutrition has been shown to negatively affect clinical outcomes, mortality, and overall healthcare expenditure in stroke survivors, it is essential to examine and monitor the nutritional status of stroke patients throughout their care journey to guide and plan, timely nutritional support and dietary modification.

Keywords: malnutrition; neurorehabilitation; nutritional status; stroke; stroke recovery; stroke rehabilitation.

Figure 1

Flowchart. Used under the Creative Commons Attribution License terms, adapted from (22).

Figure 2

(A) Prevalence of INC in the hyperacute, acute, early subacute, late subacute and chronic phase after stroke. TNE-S-E is shown in a different scale in the hyperacute, acute, and early subacute phase compared to the late subacute and chronic phase. Numbers in the plot indicate the references to the study groups and are listed below. (B) Prevalence of malnutrition in the hyperacute, acute, early subacute, late subacute and chronic phase after stroke. TNE-S-E is shown in a different scale in the hyperacute, acute, and early subacute phase compared to the late subacute and chronic phase. Numbers in the plot indicate the references to the study groups and are listed: 1, Yoo (2); 2, Gomes (3); 3, Davis (32); 4, Kokura (33); 5, Nozoe (34); 6, Nip (8); 7, Sremanakova (35); 8, Diendéré (36); 9, Vajpayee (37); 10, Gandolfo (38); 11, Crary (39); 12, NanZhu (40); 13, Zheng I (41); 14, Zheng II (41); 15, Shen (42); 16, Food Trial 2005(b) (43); 17, Xiang (44); 18, Kokura (45); 19, Barrio (46); 20, Otsuki (47); 21, Robertson (48); 22, López Espuela (49); 23, Aliasghari (50); 24, Crary (51); 25, Çoban I (52); 26, Çoban II (52); 27, Schwarz (53); 28, Porter (54); 29, Pandian (55); 30, Mosselman (56); 31, Martineau (57); 32, Ha (58); 33, Food Trial 2005(a) (59); 34, Medin (60); 35, Isono (61); 36, Far (62); 37, Brynningsen (63); 38, Kokura (64); 39, Kang (65); 40, Drozdz (66); 41, Cai (67); 42, Naito (68); 43, Hirano (69); 44, Nishioka 2020(b) (70); 45, Nishioka 2020(a) (71); 46, Kampman I (72); 47, Kampman II (72); 48, Zhang (73); 49, Shiraishi (74); 50, Hsieh (75); 51, Falsetti (76); 52, Sato (77); 53, Lim (78); 54, James (79); 55, Nishioka (80); 56, Aadal (81); 57, Aquilani (10); 58, Nishioka (82); 59, Garbagnati (83); 60, Westergren (84); 61, Poels (85); 62, Hama (86); 63, Maruyama (87); 64, Shimizu (88); 65, Carlsson (89); 66, Tsai (90); 67, Kaur (91); 68, Jung (92); 69, van Zwienen-Pot (93); 70, Campillo (94); 71, Da Silva (95); 72, Lelli (96); 73, Scrutinio (97); 74, Perry (11); 75, Vilardell (98); 76, Westergren (99); 77, Choi (100); 78, Kim (101).

Figure 3

(A) Pooled prevalence of INC in the hyperacute phase. (B) Pooled prevalence of INC in the acute phase. (C) Pooled prevalence of INC in the early subacute phase. (D) Pooled prevalence of INC in the late subacute phase. (E) Pooled prevalence of INC in the chronic phase. (F) Pooled prevalence of INC per phase.

Figure 4

(A) Pooled prevalence of malnutrition in the hyperacute phase. (B) Pooled prevalence of malnutrition in the acute phase. (C) Pooled prevalence of malnutrition in the early subacute phase. (D) Pooled prevalence of malnutrition in the late subacute phase. (E) Pooled prevalence of malnutrition in the chronic phase. (F) Pooled prevalence estimates of malnutrition per phase.

Figure 5

Prevalence of INC at follow-up. Numbers in the plot indicate the references to the study groups: 1, Yoo 2008 (2); 8, Diendéré 2020 (36); 9, Vajpayee 2008 (37); 10, Gandolfo 2019 (38); 14, Zheng 2015 II (41); 19, Barrio 2020 (46); 24, Crary 2013 (51); 30, Mosselman 2013 (56); 36, Far 2018 (62); 37, Brynningsen 2007 (63); 38, Kokura 2018 (64); 48, Zhang 2015 (73); 52, Sato 2019 (77).

Figure 6

Methods used to examine the prevalence of INC. Screening/assessment tools (S/A) (squares), anthropometrical/biochemical measurements (A/B) (circles), and other (O) (triangles) (combination of S/A tools and A/B measurements).

Figure 7

Risk of bias summary.