philpan1c

PhilSPEN Online Journal of Parenteral and Enteral Nutrition

(Article | POJ_0117.html) Issue

Original Clinical Investigation

The Prognostic Capacity of the Nutrition Risk Score and Subjective Global Assessment (SGA) Grade of the PhilSPEN Modified SGA on Mortality Outcomes.

Abstract | Introduction | Methodology | Results | Discussion | Conclusion | References | PDF (756 KB) |Back to Articles Page

Submitted: March 14, 2017 | Posted: March 18, 2018

AUTHORS:

Joyce B. Bernardino(1), MD, Luisito O. Ll ido(2), MD

  1. Fellow, Clinical Nutrition Service, St. Luke’s Medical Center- Quezon City
  2. Training Officer, Clinical Nutrition Service, St. Luke’s Medical Center- Quezon City
INSTITUTION WHERE RESEARCH WAS CONDUCTED:

Clinical Nutrition Service, St. Luke’s Medical Center, Quezon City, Metro-Manila, Philippines

ABSTRACT: | Back

Background: The Modified Subjective Global Assessment was developed for use by the Philippine Society of Enteral and Parenteral Nutrition (PhilSPEN) and is used to determine a patient's nutritional status and to stratify the patient’s nutrition risk level. Its two major components are Subjective Global Assessment and Nutrition Risk Score. There is a need to determine the capacity of this nutrition assessment tool to give prognosis of the patient's outcome

Objective: a) To determine Nutrition Risk Score and Subjective Global Assessment among patients referred to the Clinical Nutrition Services; b) To determine Outcomes such as Length of Hospital Stay, Length of ICU stay and Mortality according to Nutrition Risk Score and Subjective Global Assessment Score among patients referred to the Clinical Nutrition Services

Methodology: Retrospective study conducted at the Clinical Nutrition Services of St. Luke’s Medical Center Quezon City, a tertiary care center; subjects include all adult patients (>18 years old) that underwent nutrition assesment using the modified SGA form from January 1, 2016 to December 31, 2016. Presence of malnutrition, nutrition risk score with mortality outcomes were determined. Fisher exact test, One Way ANOVA and Odds ratio were used to determine mortality outcomes in relation to the nutritional status and different nutrition risk scores. Differences were considered statistically significant if p <0.05

Results: 787 patient records were collected; 0.89% of patients had SGA "A" status, 19.57% had SGA "B" and 79.5% had SGA "C" or severely malnourished; 40.4% had high nutrition risk levels; SGA "C" was correlated with 19.3% mortality while a high risk level is associated with 20.6% mortality; The odds that a patient with a nutrition risk score of >5 will have mortality is 2.98.

Conclusion: The modified SGA form is able to determine mortality of 19.3% when the nutritional status is SGA "C", a mortality of 20.6% when the nutrition risk level is "high", and mortality odds of 2.98 when the nutrition risk score is above 5.

KEYWORDS: SGA, Subjective Global Assessment, Nutrition Risk Score, Mortality Outcome, Assessment, prognostic

INTRODUCTION | Back

Nutrition assessment is done on hospitalized patients screened to be at risk for malnutrition. This is a more detailed evaluation to identify the level which classifies them as mild, moderate to severe risk of malnutrition and development of malnutrition related complications (1,2). This process defines the degree of urgency with which nutrition intervention should be delivered to the patient in order to correct or prevent complications. It also identifies the patients who would most likely benefit from nutrition intervention. This helps to ensure the proper route and timing for nutrition therapy and to achieve favorable outcomes (3).

There is no single clinical or laboratory indicator of comprehensive nutritional status, hence an assessment requires collection of information from a variety of sources including the historical, clinical signs/physical examination, and anthropometric, laboratory, dietary and functional domains. (4) The Modified Subjective Global Assessment was developed for use by the Philippine Society of Enteral and Parenteral Nutrition (PhilSPEN) and is used to stratify the patient’s risk level. This is comprised of the following parameters: Subjective Global Assessment, Body Mass Index, serum albumin, and total lymphocyte count.

The Subjective Global Assessment is a simple bedside method of assessing the risk of malnutrition. This was first described by Detsky. This has been demonstrated to identify malnutrition in surgical patients, those with cancer, on renal dialysis (5) and ICU (6). It is composed of the following parameters: weight loss, food intake, gastrointestinal symptoms, functional capacity, disease and relation to nutrition requirements, subcutaneous fat or muscles loss, and presence or absence of edema or ascites. An SGA of A was assigned if the patient meets all of the low risk criteria. A patient is assigned B if any of the moderate risk criteria are met or C if with any severe criteria. The Body Mass Index is one of the parameters included in the modified SGA. Anthropometric data provide an indirect assessment of body size and composition for which findings are compared to standard. (4) The Body Mass Index is interpreted according to the WHO Criteria for the Filipino population (7). Another parameter is serum albumin. This has been shown to reflect malnutrition but influenced by disease severity. It is affected by distribution due to increased albumin escape rate from the circulation associated with the cytokine response to injury and due to dilution in an increased extravascular volume. Although there is a delay in serum albumin to normal after acute illness, it may be influenced by protein and energy intake. (8) The final parameter is the Total Lymphocyte Count (TLC). Severe protein energy malnutrition is associated with a significant impairment of cell-mediated immunity, phagocyte function, complement system, secretory IgA, antibody concentrations and cytokine production. (8) Inflammatory markers such as serum albumin and total lymphocyte count improve the sensitivity/specificity of indexes predicting the risk of malfunction. Inflammatory activity negatively influences mobility, longevity, infectious complications, in chronic and acute disease conditions (1)

The study by Lacuesta-Corro et al. showed that the modified SGA nutritional assessment form had a very good ability to determine nutritional status and risk level determination with a Sensitivity of 94.7%, Specificity of 95.2%, Positive Predictive Value of 95.7%, Negative Predictive Value at 94.1%, and Diagnostic Accuracy of 95%.  (9) The components of the Modified SGA form include the SGA grade and the Nutrition Risk Score. The Nutrition Risk score is the weighted sum of the 4 parameters (SGA, BMI, Albumin and TLC). The aim of this study is to correlate the SGA grade and Nutrition Risk Score with mortality outcomes and to establish the capacity of this tool to predict outcomes such as mortality or length of hospital stay.

The objectives of the study are:

  1. To determine Nutrition Risk Score and Subjective Global Assessment among patients referred to the Clinical Nutrition Services
  2. To determine Outcomes such as Length of Hospital Stay, Length of ICU stay and Mortality according to Nutrition Risk Score and Subjective Global Assessment Score among patients referred to the Clinical Nutrition Services

METHODOLOGY | Back

Setting and Study design:
This is a retrospective study conducted at the Clinical Nutrition Services of St. Luke’s Medical Center Quezon City, a tertiary care center. Upon admission, patients undergo nutrition screening using the NRS 2002, and once seen or referred to the Clinical Nutrition Service undergoes Nutrition Assessment using the PhilSPEN Modified SGA, which is the standard nutritional assessment tool of the institution .

Selection of patients:
Subjects include all adult patients (>18 years old) that underwent nutrition assessment from January 1, 2016 to December 31, 2016 by the Clinical Nutrition Service Fellows. The cases are mixed - medical and/or surgical. A total of 800 referrals and assessments were seen by the Clinical Nutrition Service Fellows. Included in the study were 787 charts (98.37 %). Excluded in the study are patients still admitted at the time of data collection (January 2017) since outcome cannot be determined at the time. Pregnant/Postpartum patients were also excluded, since the BMI will be unreliable.

Collection of Data:
The Modified Subjective Global Assessment Tool by PhilSPEN was retrieved from patient’s charts, which is already collected as part of routine patient care. The values for weight, albumin, total lymphocyte count used were those available at the time of assessment. Nutrition risk score is categorized as follows: Score of 0 to 2: Low Risk, Score of 3 to 5: Moderate Risk, and Score of 6 to 9: High Risk. The outcomes collected include status upon discharge whether dead or alive, length of hospital stay (all patients seen), time to discharge alive (e xcluding expired patients) .

Statistical Analysis
The data was organized in Excel software. The statistical measures taken were proportions (percent), mean, median, standard deviation, and odds ratio. For the mortality, Fisher exact test was done between low and moderate risk, low to high risk, and moderate to high risk patients. For length of hospital stay, One Way ANOVA was done between low, moderate and high risk. Odds ratio between nutrition risk score and nutrition score cut-off was computed. Differences were considered statistically significant if p <0.05. The NCSS v10 statistics software was used.

Ethics
As a retrospective stud which used available data collected as part of routine patient care, consent was not required, ethics committee approval was not done.

RESULTS | Back

Patient Profile Data:

sgatbl01

For the patient characteristics, majority of patients seen according to Nutrition Risk Score are Moderate Risk (47.5%), followed by High Risk (40.4%) and a small percentage are Low Risk (12%). Most of the patients seen are new referrals (89%) compared to reassessments (10.5%). There are more Male (56%) than Female (44%) patients. Most of the patients that are seen are admitted in the Critical Care Units (58.5%) compared to the Regular Rooms (41.4%). Majority of patients are in the geriatric population (>60 years old) (61.88%).

sgatbl02

Using the Modified Subjective Global Assessment, most of the patients have a Subjective Global Assesment Grade of C (79.5%) followed by B (19.5%) and with a small group falling under A (0.89%).  According to BMI, majority of patients are normal (50.83%) followed by undeweight/obese patients (25.41%) and overweight (23.76%).

Majority of patients do not have an albumin determination at the time of assessment at (42.95%) but among patients who have, most are moderately decreased at 2.5-3.4 g/dL (21.86%), followed by those with severely decreased albumin at <2.5 g/dl (21.86%). Only a small number has normal albumin determination. (6.48%).

For the total lymphocyte count, most patients have a normal value >1500 cells/mm3 (41.04%) followed by those with severely decreased TLC (29.98%) and those with moderately decreased TLC (28.97%).

OUTCOMES:

sgatbl03

Majority of patients have a Subjective Global Assessment of C (severe), at 79.54%, followed by SGA B (moderate) at 19.57% and least of all is SGA A (low risk) at 0.89%.  There is an increasing mortality rate, with Low Risk patients having no mortalities, while SGA B has 1.95% mortality, and the highest is among SGA C patients at 19.33%.

sgatbl04

Among patients referred to the Clinical Nutrition Services, those with Low Risk Score of 0-2 did not have any mortalities. Those who were classified as Moderate Risk has a mortality of 12.29% while High Risk patients have the highest mortality of 20.63%. This is statistically significant between groups.

sgatbl05

Comparing patients with a low to moderate nutrition risk score, high risk patients have an odds ratio of 2.99 risk of mortality, which is significantly increased among groups.

sgatbl06

Low risk patients have a shorter hospital length of hospital stay, with a mean of 6.94 and median of 5 days from time of admission. This is increased among Moderate risk patients with a mean of 16.44 days, median of 11 days. The longest length of hospital stay is among High risk patients with a mean of 26.29 days and median of 16. One way ANOVA was done between groups shows that this is statistically significant.

sgatbl07

Excluding patients with mortality among those who are discharged alive, low risk patients have a shorter length of stay with an average of 6.94 days and median of 5. This is increased among moderate risk patients with mean of 15.87 days, median of 11 days. The longest time to discharge alive is among high risk patients with a mean of 24.69 days and median of 15 days.

sga_tbl08

Approximately 20.97% of patients are seen on the first day of admission. The range between referral and assessment is between zero to 14 days for low risk patients, up to 52 days for moderate risk, and up to 64 days for high risk patients. There is also an increasing mean average days and standard deviation.

More than half of patients are referred until discharge (52.48%), the rest (47.52%) are signed out (high risk package, comprehensive geriatric assessment, patient or physician’s preference)

DISCUSSION: | Back

In this study, majority of patients seen by the SLMC Clinical Nutrition Services (CNS) according to Nutrition Risk Score are Moderate to High Risk. Using the Subjective Global Assessment, most patients have an SGA Grade of C. This may be due to the fact that referred patients are likely to have impaired nutrition related symptoms or laboratory parameters. Majority of patients seen are admitted to the critical care units and this may be due to the hospital policy of automatic CNS assessment upon ICU admission. A study by Singh shows that in the hospital ward, majority of patients are SGA B (39%) followed by A (31%) then C( 30%) (10). Similarly, in a study done by Kwang, in palliative care unit of a Malaysian hospital, 50% were SGA B, followed by 31% with SGA A, and 19% with SGA C. (11). At the ICU, a study by Fontes, shows that 41.6% are moderately malnourished, and 12.4% are severely malnourished. (6)

Using the PhilSPEN Modified SGA, the study showed that mortality increases with higher Nutrition Risk Score as a combination of SGA, BMI, serum albumin and total lymphocyte count. It was able show mortality at 12.29% when the Nutrition Risk Score is from 4 to 6 (moderate) and 20.63% when the nutrition risk score is at 7 to 9 (high). Using odds ratio, there is 2.99 increased mortality between those with a nutrition risk score >6 versus those with a score <5.  

The first study on outcome using the PhilSPEN modified SGA was in surgical patients which showed no mortality for low- moderate risk patients, but had 18.2% mortality in patients classified as High Risk, with a nutrition risk score >4. There was also increase in complication rate and hospital stay. (12)

Using the Subjective Global Assessment, it was demonstrated that mortality increases with SGA grade. This may be due to the known complications of malnutrition, regardless of the underlying condition. A study by Bector, shows that among medical ICU patients, mortality rates were higher according to SGA with 10.8% among SGA A, 45.5% among SGA B and 55.6% among SGA C. (15)A study by Ferreira in liver transplant patients, showed that those who were malnourished using SGA had a higher mortality at 39.1% compared to the overall population at 20.7%. (13)Meanwhile, a study among cancer patients showed that those with SGA A had a longer survival of 19.3 months, compared to those with SGA B with 15.5 month and SGA C at 6.7 months. (14)

Regarding other parameters, most of the patients seen have a normal BMI, including high risk patients. The underestimation of malnutrition with BMI,(16) may be despite significant weight loss, if there is a pre-diagnosis obesity or overweight status. (11) The method of determination, may also be ascertained since not all patients are re-weighed at time of assessment, some due to technical difficulties.

In this study, not all patients have an albumin determination. Similarly, a study in the University of Manitoba shows that serum albumin is found in only 62% of patients upon admission. (10) This may be since it is not routinely done, especially in low risk patients, or those without a renal or hepatic pathology.

Most patients have a normal total lymphocyte count upon assessment. One study by Ozbilgin showed that white cell count was not a significant predictor of outcome in the post operative care unit (17), while another study by Bector among Medical ICU showed that lower WBC correlated with mortality. (15)

Length of hospital stay increases with nutrition risk score as well as time to discharge alive. In a study by Tsaousi in a Greek hospital, among the paramenters that increased length of hospital stay are weight loss and loss of appetite, (16) both found in the SGA. Another study among patients in a post operative care unit, SGA Grade has a significant correlation with length of stay. (17)

This study further strengthens the value of the PhilSPEN modified SGA as a nutritional assessment tool and a good prognostic tool for mortality and length of hospital stay.

CONCLUSION: | Back

The modified SGA form is able to determine mortality of 19.3% when the nutritional status is SGA "C", a mortality of 20.6% when the nutrition risk level is "high", and mortality odds of 2.98 when the nutrition risk score is above 5.

REFERENCES: | Back

  1. Soeters, PB, Reijven PL et al. A rational approach to nutrition assessment. Clinical Nutrition 2008; 27: 706-16.
  2. Mueller C, Compher C, American Society for Parenteral and Enteral Nutrition (A.S.P.E.N) Board of Directors. A.S.P.E.N. Clinical Guidelines. Nutrition screening, assessment, and intervention in adults. J Parenter Enteral Nutr 2011; 35; 16-24.
  3. Seres DS. Nutrition Support for the Critically Ill. 2016. Humana Press.
  4. Mueller, Charles, Stephen McClave, et al. The A.S.P.E.N. Adult Nutrition Support Core Curriculum. 2nd ed. 2012. Print.
  5. Yang, Fwu-Lin et al. "A Cohort Study Of Subjective Global Assessment And Mortality In Taiwanese Hemodialysis Patients". Renal Failure 29.8 (2007): 997-1001. Web.
  6. Fontes, Daniel, Simone de Vasconcelos Generoso, and Maria Isabel Toulson Davisson Correia. "Subjective Global Assessment: A Reliable Nutritional Assessment Tool To Predict Outcomes In Critically Ill Patients". Clinical Nutrition 33.2 (2014): 291-295. Web.
  7. Llido, LO et al. Comparison of the Body Mass Index based nutritional status using WHO criteria versis “Asian” Criteria: report from the Philippines, 2011 (http://dpsys120991.com/Philspen_Online_Abstracts4.php)
  8. Sobotka, Luboš and Simon Philip Allison. Basics In Clinical Nutrition. Fourth ed. Prague: Galén, 2011. Print.
  9. Lacuesta-Corro L and Llido LO.  The results of the validation process of a Modified SGA (Subjective Global Assessment) Nutrition Assessment and Risk Level Tool designed by the Clinical Nutrition Service of St. Luke’s Medical Center, a tertiary care hospital in the Philippines. PhilSPEN Online Journal of Parenteral and Enteral Nutrition; (Article 12 | POJ_0002.html) Issue February 2012 - December 2014: 1-7. Available at http://dpsys120991.com/POJ_0002.html. Accessed September 24, 2016.
  10. Singh, Harminder et al. "Malnutrition Is Prevalent In Hospitalized Medical Patients: Are Housestaff Identifying The Malnourished Patient?". Nutrition 22.4 (2006): 350-354. Web.
  11. Ang Yee Kwang, and M. Kandiah. "Objective And Subjective Nutritional Assessment Of Patients With Cancer In Palliative Care". American Journal of Hospice and Palliative Medicine 27.2 (2009): 117-126. Web.
  12. Ocampo et al. Predicting Post-operative Complications Based on Surgical Nutritional Risk Level using the SNRAF in Colon Cancer Patients: A Chinese General Hospital & Medical Center Experience. . PhilSPEN Online Journal of Parenteral and Enteral Nutrition; (Article 7 | POJ_0012.html) Issue January 2010 - January 2012: 55-66. Accessed September 24, 2016. Available at http://dpsys120991.com/POJ_0012.html
  13. Ferreira, L.G. et al. " SEVERE MALNUTRITION IS A PREDICTOR OF MORTALITY IN PATIENTS ON THE WAITING LIST FOR LIVER TRANSPLANTATION". Clinical Nutrition Supplements 5.2 (2010): 132. Web.
  14. Gupta, Digant et al. "Can Subjective Global Assessment Of Nutritional Status Predict Survival In Ovarian Cancer?". Journal of Ovarian Research 1.1 (2008): 5. Web.
  15. Bector, Savita et al. "Does The Subjective Global Assessment Predict Outcome In Critically Ill Medical Patients?". Journal of Intensive Care Medicine 31.7 (2016): 485-489. Web.
  16. Tsaousi, Georgia et al. "Prognostic Indices Of Poor Nutritional Status And Their Impact On Prolonged Hospital Stay In A Greek University Hospital". BioMed Research International 2014 (2014): 1-8. Web.
  17. Özbilgin, Şule et al. "Morbidity And Mortality Predictivity Of Nutritional Assessment Tools In The Postoperative Care Unit". Medicine 95.40 (2016): e5038. Web.
  18. Dawson B,  Trapp R. Basic & Clinical Biostatistics. 2004. Lange Medical Books/McGraw-Hill 2004
  19. Detsky AS, McLaughlin JR, Baker JP et al. What is Subjective Global Assessment of Nutritional status? J Parenter Enteral Nutr 1987; 11(1): 8-13
     

Abstract | Introduction | Methodology | Results | Discussion | References | Back to Articles Page