This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Oh, H. Articles by Seo, W. Search for Related Content PubMed PubMed Citation Articles by Oh, H. Articles by Seo, W. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Medline Plus Health Information Fluid and Electrolyte Balance Stroke Hazardous Substances DB POTASSIUM SODIUM Social Bookmarking                   What's this?

Age Differences in Fluid Balance and Serum Na⁺ and K⁺ Levels After Nasogastric Tube Feeding in Stroke Patients: Elderly vs Nonelderly

From the Department of Nursing, College of Medicine, Inha University, Incheon, Republic of Korea

Correspondence: WhaSook Seo, PhD, Professor, Department of Nursing, College of Medicine, Inha University, YongHyun Dong 253, Incheon, 402-751, Republic of Korea. Electronic mail may be sent to wschang{at}inha.ac.kr.

Background: Enteral nutrition support has been commonly used to improve nutrition status in acute stroke patients. The purpose of this study was to examine whether significant alterations in fluid and serum Na⁺ and K⁺ levels due to nasogastric tube feeding depended on patient age. Methods: This study was conducted by retrospective review of the medical records of tube-fed patients with acute brain infarction who had been hospitalized in a university hospital. Results: Rates of overhydration were high before tube feeding in both elderly and nonelderly stroke patients. This overhydration rate became more prevalent after iso-osmolal tube feeding in the elderly but decreased somewhat in the nonelderly subjects, and this difference in fluid balance after tube feeding between the elderly and nonelderly was statistically significant. Contrary to fluid balance, serum Na⁺ and K⁺ concentrations were no different in the elderly and nonelderly patients after tube feeding. Conclusions: Although the results showed that overhydration state was prominent after tube feeding in the elderly, it appeared that there was a very mild clinical impact because the osmolality and serum sodium were not changed in this group. However, more attention to the maintenance of fluid balance in the elderly tube-fed stroke patients is needed because elderly patients have great difficulty in maintaining a normal fluid balance.

Acute stroke patients can be complicated by dysphagia in about 25%–42% of cases due to weaknesses of the oral musculature and tongue movements or reduced pharyngeal receptor sensitivity.^1,2 Moreover, drugs that are commonly used for stroke management may also precipitate or aggravate dysphagia.² It is known that dysphagia in stoke is usually transient and that more than half of stroke patients show improvement by the end of the first week.³ However, the symptoms persist in about 8% of patients for 6 months or more.⁴

When dysphagia continues for a protracted time, it can lead to several nutrition and medical problems, such as malnutrition, inadequate hydration, aspiration pneumonia, airway obstruction, and irritation of the mucus membrane.^5,6 Studies have shown that the prevalence of malnutrition increases from 16% on admission to 22%–35% about 2 weeks after admission.^3,7 In addition, malnutrition increases complication rates and is associated with a poorer functional outcome, a prolonged hospital stay, and a reduced quality of life.^8–10 However, malnutrition may receive little attention in acute stroke patients due to other serious problems.

Enteral nutrition support has been commonly used to improve nutrition status in acute stroke patients.¹¹ This technique is known to be more economical and physiologic, and to induce fewer complications than parenteral nutrition (PN) support.¹² Several enteral nutrition methods have been proposed (ie, nasogastric, surgical gastrostomy, percutaneous gastrostomy, and jejunostomy). Of these, nasogastric tube feeding is most commonly used in clinics because of the following advantages: avoidance of surgery and general anesthetics, a shorter procedure, and a lower cost.¹³

However, enteral nutrition therapy can also cause several complications (ie, gastrointestinal [nausea, vomiting, diarrhea, constipation, and reflux], mechanical [transnasal catheter dislocation or obstruction, lower esophageal sphincter incompetence, and ulceration of nasal mucosa], and metabolic [fluid and electrolyte imbalance].¹⁴ Of these complications, gastrointestinal, respiratory, and mechanical complications have been reduced by recent technologic advances in enteral delivery systems. However, associated fluid and electrolyte imbalances continue to exist. In particular, this complication can be serious in brain-damaged patients with altered consciousness.

Most frequently, the reported fluid and electrolyte imbalances associated with nasogastric tube feeding are dehydration, hypernatremia, hyponatremia, and hyperkalemia.^14–18 According to Kubo et al,¹⁹ the tubefed patients that are prone to fluid and electrolyte imbalance are those with renal and cardiac diseases, elderly patients, and patients with neurohypophyseal damage.

Obviously, age is a significant variable to consider in relation to the incidence of tube-feeding complications; however, few studies have been conducted on this matter. In addition, many of the previous studies are in the form of case reports, literature reviews, or descriptive staged studies with uncontrolled extrinsic variables (ie, including various disease groups). Therefore, a more systematic analysis, excluding the influence of other extrinsic variables and with a large sample size, was needed. The purpose of the present study was to examine whether significant alterations in fluid and serum Na⁺ and K⁺ levels due to nasogastric tube feeding are significantly dependent on age.

	MATERIALS AND METHODS

Top MATERIALS AND METHODS RESULTS DISCUSSION

 
Research Design and Subjects
The present study was conducted according to a retrospective review of the medical records of tube-fed patients with acute brain infarction who had been hospitalized in a university hospital. The data collection was limited to 1 university hospital to eliminate possible variations in feeding method and the formula types used.

Only patients with cerebral infarction were included in the present study, for following reasons: (1) to reduce the influences of disease-related factors, (2) to minimize variations due to pharmacologic treatments, because acute brain infarction patients are provided with a relatively fixed medication regimen compared with hemorrhagic stroke patients, and therefore, the pharmacologic treatment administered before tube feeding was similar to that after tube feeding.

The subjects with serum creatinine >2 or abnormal serum glutamic-oxaloacetic transaminase (SGOT)/serum glutamic-pyruvic transaminase (SGPT) shown before tube feeding were excluded because alterations in renal or hepatic functions might affect hydration status or serum electrolyte balance. Finally, 76 subjects were included in the present study.

Data Collection
For this present study, the medical records of 76 subjects with all necessary data were reviewed retrospectively, with the permission of the responsible institutional review board. Data related to fluid balance (intake/output) and serum Na⁺ and K⁺ levels on 4 consecutive days (from the day before to third day of tube feeding) were collected. To confirm that alterations in outcome variables (fluid balance and serum Na⁺ and K⁺ levels) were produced by tube feeding and not by alterations in extrinsic variables (eg, medication or changes in the functions of fluid/electrolyte-related organs), the following data were also collected: types and doses of the prescribed medications, serum creatinine, SGOT, and SGPT. From the medication records, prevalently applied infusions, transfusions, and oral medications for acute brain infarction patients were listed by reviewing the medical records of 20 randomly selected subjects in advance. Only those medications known to affect fluid and electrolyte levels were selected with the aid of a pharmacology professor. Medication data sheets were prepared for data collection.

Measurements
The following data were recorded: gender, age, diagnosis, types and doses of medications administered before and after tube feeding, types and calories of formula, serum Na⁺ and K⁺ levels, serum creatinine, SGOT, SGPT, blood glucose level, blood osmolality, and 24-hour fluid input and output (I/O). The 24-hour intake included all fluids (ie, enteral, oral, IV infusion, and PN).

Data Analysis
Data analysis was performed using SPSS/PC 12.0 for Windows. To test the homogeneities of the types and dosages of medications and feeding calories before and after tube feeding or between the elderly and nonelderly group, repeated-measure multivariate analysis of variance (RM-MANOVA) or analysis of variance (RM-ANOVA) was used. In general, RM-MANOVA or -ANOVA can be used to determine whether there exist statistically significant differences among data measured repeatedly in a same group. For repeatedly measured data, MANOVA or ANOVA cannot be applied, because of the violation on a basic assumption (ie, the measurements are independent of each other).

To test the homogeneity of renal (using blood urea nitrogen [BUN] and serum creatinine levels) and hepatic functions (using SGOT and SGPT levels) after tube feeding between the elderly and nonelderly group, ² test was used. To compare the proportions (in percentage terms) of high, normal, and low fluid balances and serum Na⁺ and K⁺ levels in the elderly and nonelderly groups, ² test was also used. The ² test is a statistical test commonly used to examine goodness of fit, homogeneity, or independence for the categorical variables by comparing actual (observed) numbers with those that would be expected according to theory or simply by chance.

Study Limitations
In general, the most usual way to approach a project like the present study is to exclude those with confounding disease, such as diabetes or renal or hepatic dysfunction. Accordingly, those with serum creatinine >2 (n = 3) or abnormal SGOT/SGPT (n = 6) shown before tube feeding were excluded from the present study. However, subjects with diabetes could not be excluded because there were too many of them, resulting in a possible compounding influence.

Another study limitation is that the 24-hour I/O measurement used in the present study is a weak indicator prone to measurement error. Data of body weight, a more stable indicator of fluid balance, could not be obtained in the present study, because the body weights of stroke patients were measured mostly once a week in the institution in which data were collected. This is one of the limitations of retrospective data retrieval. The present study also has a limitation in the ability of general application due to a small sample size (n = 76).

	RESULTS

Top MATERIALS AND METHODS RESULTS DISCUSSION

 
Descriptive Analysis of General and Illness-Related Characteristics
The medical records of 76 subjects with all necessary data were included. According to the results of ground of related studies, elderly was defined as those older than 66 years.^20,21 To examine age differences, subjects were divided into an elderly (n = 37) and a nonelderly group (n = 39).

Mean ages were 77.0 (±6.2; range: 67–90) and 59.0 (±8.3; range: 38–66) years in the elderly and nonelderly groups, respectively (Table I). Fourteen (37.8%) were men and 23 (62.2%) women in the elderly group, and 27 (69.2%) were men and 12 (30.8%) women in the nonelderly group. In both groups, the most prevalent infarction site was the cerebral cortex (78.4% of the elderly; 79.4% of the nonelderly), followed by the brain stem and cerebellum. Subjects started feeding through a nasogastric tube within 1–3 days after admission (first day 44.7%, second day 31.6%, and third day 23.7%), and 96% were provided an iso-osmolal feeding formula (1 kcal/mL).

Descriptive Analysis of the Major Variables
Mean fluid balance over 24 hours (subtracting output from intake volume) in the elderly group was within the normal range (–500 to 500 mL) on the day before (37.47 ± 189.05) and on the first days (476.65 ± 132.95) of tube feeding but increased to above the normal range on the second (609.69 ± 101.05) and third days (597.96 ± 194.45) of tube feeding (in mL). On the other hand, all the mean values of fluid balance in the nonelderly group were maintained within the normal range (55.08 ± 167.57, 448.83 ± 174.43, 489.00 ± 184.22, and 196.08 ± 179.62 on the day before and on the first, second, and third days of tube feeding, respectively, all quoted in mL, Table II).

Serum Na⁺ levels on 4 consecutive days (from the day before until the third day of tube feeding in sequence) were 138.42 (±4.81), 139.14 (±4.35), 139.47 (±4.75), and 138.24 (±8.86) in the elderly group and 140.18 (±6.41), 141.52 (±7.00), 141.47 (±9.04), and 142.57 (±11.13) in the nonelderly group, indicating that all were in the normal range (136–145, in mEq/L, Table II). All serum K⁺ levels on these 4 consecutive days were also maintained within the normal range of 3.5–5.0 mEq/L (3.83 ± 0.53, 3.69 ± 0.39, 3.77 ± 0.66, and 3.79 ± 0.66 in the elderly and 3.89 ± 0.50, 3.95 ± 0.68, 3.85 ± 0.55, and 3.81 ± 0.56 in the nonelderly, in mEq/L, Table II).

Serum glucose on these 4 consecutive days in sequence were 162.83 (±60.04), 153.86 (±47.11), 150.31 (±63.59), and 150.25 (±64.32) in the elderly group and 169.38 (±111.35), 142.00 (±48.93), 155.45 (±77.78), and 141.72 (±59.88) in the nonelderly group, meaning that all were above normal (all quoted in mg/dL, Table II). On the other hand, all values of serum osmolality were in the normal range: 308.45 (±17.81), 302.27 (±17.41), 302.30 (±17.08), and 300.64 (±15.35) in the elderly and 293.05 (±10.11), 293.89 (±17.40), 308.56 (±52.99), and 295.80 (±21.32) in the nonelderly (all quoted in mOsm, Table II).

Medications That Can Affect Hydration Status or Electrolyte Balances
Of medications that can affect subjects' hydration or electrolyte levels, the prevalent IV infusions given to subjects were normal saline (given to 79.5%–94.6% of the subjects), mannitol (37.8%–48.6%), dextran 40 (7.7%–21.6%), Frucenil (Dong Shin, Seoul, Korea) or Cerol (Choongwae Parma Corporation, Seoul, Korea; both 6.1%–28.2%), 20% albumin (5.4%–13.5%), 5% dextrose in sodium chloride (5.4%–12.8%), and 5% dextrose in water (0%–13.3%), and less frequently 10% FreAmine (Penang, Maylasia, 2.4%–4.7%). The injection medications given to the subjects were furosemide, KCl, dexamethasone, and NaCl. Of those, furosemide (10.7%–32.4%) and KCl (12.8–32.4%) were most frequently provided (Table III).

It appears that there exist no differences in the medication types given to subjects between the elderly and nonelderly group, as well as before and after tube feeding. As an exception, however, the percentages of the subjects taking Lasix increased approximately twice after tube feeding in both elderly and nonelderly groups (Table III).

Homogeneity Testing for Extrinsic Variables
Homogeneity in medication, feeding calories, serum creatinine, SGOT, and SGPT before and after tube feeding was examined to confirm that alteration in outcome variables was produced by tube feeding and not by alterations in medications, feeding calories, or hepatic or renal function (Table IV).

As mentioned above, the subjects with abnormal serum creatinine or SGOT/SGPT shown before tube feeding were already excluded from the present study. Therefore, only whether there was a significant difference in the proportions of the subjects with abnormal serum creatinine (>2.0) or SGOP/SGPT after tube feeding between the elderly and nonelderly group was examined in this analysis. Results showed that the proportions of the subjects with abnormal creatinine after tube feeding were homogeneous between the groups (first day ² = 1.48, p = .49; second day ² = 1.47, p = .49; third day ² = 0.02, p = 1.00, Table IV). The proportions of the subjects with abnormal SGOT/SGPT after tube feeding were also homogeneous between the groups (first day ² = 0.12, p = 1.00; second day ² = 0.03, p = 1.00; third day ² = 0.29, p = .67 for SGOP; first day ² = 0.77, p = .56; second day ² = 0.10, p = 1.00; third day ² = 0.13, p = 1.00 for SGPT, Table IV). These results indicate that renal or hepatic function did not contribute to any alteration in fluid or electrolyte levels after tube feeding.

With respect to medications, the types and doses of infusions (Wilks = 0.80, F = 1.47, p = .16) and injection and oral medication (Wilks = 0.89, F = 1.01, p = .41) were homogeneous throughout tube feeding in both groups. In addition, the tube feeding calories provided to the subjects were also homogeneous between the groups (Table IV).

Comparisons of Changes in Fluid Balance, and Serum Na⁺ and K⁺ Levels Before/After Tube Feedings in Elderly vs Nonelderly Subjects
For this analysis, subjects were divided into 3 groups according to the normal values of fluid balance, and serum Na⁺ and K⁺ concentrations (high, normal, and low) and the proportion of subjects in each group were calculated as percentages.

Twenty-Four-Hour Fluid I/O
Before tube feeding, low 24-hour fluid I/O (<500 mL) was found in 20.6% and 16.7%, normal (–500–500 mL) in 55.9% and 50.0%, and high (>500 mL) in 23.5% and 33.3% of the elderly and nonelderly subjects, respectively (Table V, Figure 1A, B). However, no significant differences were found in the proportions of over-, normal, or underhydration in the elderly and nonelderly subjects before tube feeding (² = 0.85, p = .25), and this tendency continued until the second day of feeding (first day ² = 0.18, p = .52; second day ² = 3.29, p = .08).

View larger version (21K): [in this window] [in a new window]   FIGURE 1. Fluid balance and serum sodium and potassium levels before and after tube feeding. Changes in fluid balance in elderly (A) and nonelderly (B). Changes in serum sodium levels in elderly (C) and nonelderly (D). Changes in serum potassium levels in elderly (E) and nonelderly (F).

On the third day of tube feeding, these proportions became significantly different between the groups (² = 4.60, p = .04): underhydration was observed in 0% and 11.5%, normal hydration in 50.0% and 57.7%, and overhydration in 50.0% and 30.8% of the elderly and nonelderly subjects, respectively. One point worth noting is that overhydration rate on the third day in the elderly group was relatively high, 50.0% as compared with 30.8% in the nonelderly.

The proportion with normal hydration before tube feeding was 55.9% in the elderly, and this proportion decreased slightly to 50.0% after tube feeding. At the same time, underhydration rates decreased (from 20.6% to 0.0%), but overhydration rates were increased (from 23.5% to 50.0%) after tube feeding in the elderly (Table V, Figure 1A). This result signifies that elderly subjects with iso-osmolal tube feeding tend to overhydrate during tube feeding.

On the contrary, the proportions of each hydration status in the nonelderly subjects did not markedly change (Table V, Figure 1B). In this group, the proportions of under- and overhydration slightly decreased after tube feeding (16.7%11.5% for underhydration, and 33.3%30.8% for overhydration), but those of normal hydration slightly increased after tube feeding (50.0%57.7%).

Serum Na⁺ levels
Before tube feeding, hyponatremia was observed in 28.1% and 23.5%, normal serum Na⁺ level (136–145 mEq/dL) in 68.8% and 61.8%, and hypernatremia in 3.1% and 14.7% of the elderly and nonelderly subjects, respectively (Table V, Figure 1C, D). However, no significant differences were found between the two groups with respect to these proportions (² = 2.69, p = .18, Table V), and this tendency continued after tube feeding (first day ² = 2.49, p = .32; second day ² = 1.71, p = .17; third day ² = 0.74, p = .29, Table V). Notably high incidence rates of hyponatremia were observed before tube feeding in both groups.

Before tube feeding, normal serum Na⁺ levels were observed in 68.8% and 61.8% of the elderly and nonelderly subjects, respectively. These proportions increased after tube feeding in both groups; however, increments were more obvious in the nonelderly (61.8%75.0%) than in the elderly (68.8%76.0%). Such increments appeared to be due to the decrements in the rate of hyponatremia. However, the incidence rates of hypernatremia were not noticeably changed by tube feeding in either group.

Serum K⁺ Levels
Before tube feeding, hypokalemia was observed in 18.8% and 17.6%, normal serum K⁺ levels (3.5–5.0 mEq/dL) in 78.1% and 79.4%, and hyperkalemia in 3.1% and 2.9% of the elderly and nonelderly subjects, respectively. However, no significant differences in these proportions were found between the 2 groups (² = 0.02, p = .58, Table V), and this tendency continued after tube feeding (first day, ² = 2.01, p = .37; second day, ² = 0.08, p = .50; third day, ² = 1.98, p = .17, Table V).

Hypokalemia was more prevalent than hyperkalemia before and after tube feeding in both groups. The proportions of hyperkalemia and hypokalemia were not markedly changed after tube feeding in the elderly group, whereas those of hypokalemia increased (17.6%32.1%) and normal serum K⁺ levels decreased (79.4%67.9%) in the nonelderly group (Table V, Figure 1E, K).

Comparisons of Serum Glucose and Na⁺ Levels Between Diabetic and Nondiabetic Subjects
In the present study, the diabetic patients represent 24% of study group and a greater proportion of the elderly than the nonelderly (Table I). This factor might result in the high incidence of hyperglycemia shown in Table V. To clarify this matter, the comparisons of serum glucose levels in the diabetic vs nondiabetic subjects were conducted in the present study. In addition, serum Na⁺ level was also compared in the diabetic vs nondiabetic subjects.

Results showed that no significant differences in the proportions of hyper-, normo-, and hypoglycemia were found between the diabetic and nondiabetic subjects in both elderly and nonelderly groups (Table VI). Similarly, there were no significant differences in the proportions of hyper-, normo-, and hyponatremia between the diabetic and nondiabetic subjects in both elderly and nonelderly groups (Table VI).

	DISCUSSION

Top MATERIALS AND METHODS RESULTS DISCUSSION

 
The results obtained during this study show that elderly and nonelderly subjects are not significantly different in terms of the proportions of under-, normo-, or overhydration before tube feeding but that these proportions became significantly different on the third day of tube feeding. Before tube feeding, a normal fluid balance was observed in 53% of all subjects, and the incidence in overhydration (28.4%) was higher than the incidence of underhydration (18.7%). This overhydration tendency became more prominent after tube feeding and rose 50% by the third day of tube feeding in the elderly group. On the other hand, the overhydration tendency was rather decreased slightly (33.3%30.8%) in the nonelderly, and more subjects fell into the normal fluid balance range after tube feeding.

Although overhydration tendency became more prominent after tube feeding in the elderly group, this appeared to exert a very mild clinical impact because the osmolality and serum sodium levels were found to be unchanged in this group. Moreover, I/O balance itself was not a strong indicator of fluid balance. Accordingly, a more stable and accurate measurement tool of fluid balance, such as body weight, was needed. However, body weights of stroke patients were measured mostly once a week in the institution in which data were collected unless there was a special reason. Further studies on this matter are needed to obtain more reliable findings.

Hyperosmolar feeding formulas have been frequently used to achieve high-protein intake in the past. However, it has been persistently reported that hyperosmolar feeding may cause dehydration and hypernatremia development, especially in the elderly, because they have a reduced renal concentrating ability and thus excrete more water in urine.¹⁸ Recently, iso-osmolar feeding formulas have been favored by clinics. Most subjects in the present study were also provided iso-osmolar feeding formulas, and this probably contributed to the lower incidences of dehydration and hypernatremia observed.

As discussed above, approximately 28% of all subjects had an overhydrated status at baseline, although the degree of overhydration was relatively mild in both groups. This high incidence rate of overhydration in stroke patients might be due to increased ADH secretion, which is frequently observed in stroke patients due to brain injury or to the adoption of a continuous supine position.^18,21 Moreover, stroke patients commonly receive medications favoring sodium loss and water retention in the acute stage.

Taken together, it is evident that elderly tube-fed stroke patients are at particular risk of developing a fluid imbalance whether fed hyper-, iso-, or hypo-osmolal formulas. The potential for developing dehydration with hyperosmolal formulas is clearly illustrated by other previous studies, in addition to the potential for overhydration with hypo-osmolal formulas.²¹ Furthermore, our results indicate that iso-osmolal formulas can induce overhydration in elderly stroke patients. Nevertheless, the negative effect of iso-osmolar tube feeding on the fluid balance of the elderly shown in the present study cannot be deterministically concluded due to the limitation of the indicator of fluid balance, as mentioned above. Besides, the serum osmolality and sodium levels were found to be unchanged in this group. It can be inferred that the overhydration tendency shown after tube feeding in the elderly appears to exert a very mild clinical impact.

Contrary to fluid balance, in the present study, serum Na⁺ and K⁺ concentrations were no different in the elderly and nonelderly patients after tube feeding. Before tube feeding, hyponatremia was observed in 28.1% and 23.5% of the elderly and nonelderly subjects, respectively, and was more prevalent than hypernatremia. The high incidence rates of hyponatremia at baseline in the 2 groups can be explained by the same reasons given for overhydration above (ie, increased anti-diuretic hormone (ADH) secretion—diluted hyponatremia—and medications favoring sodium loss in stroke patients). However, the incidence rates of hyponatremia were decreased by about 10% in both groups after tube feeding, whereas the incidences of hypernatremia were not noticeably altered.

Most early studies on the complications of tube feeding focused on hypernatremia because, at that time, hyperosmolal feeding formulas were generally used to provide high protein.²¹ However, given the increased popularity of iso-osmolal formulas, hyponatremia has become a major problem. Moreover, many clinicians dilute formulas to hypotonic levels during the first few days of tube feeding, and this also results in hyponatremia. In the present study, all subjects received isoosmolal formulas without any dilution, even at the beginning of tube feeding. Therefore, the high incidences of hyponatremia observed before and after tube feeding can be associated with the pathologic process of stroke, as described above, and not with tube feeding.

On the other hand, hyperglycemia can induce either hyponatremia (called pseudohyponatremia) or hypernatremia (by diuresis). Because the majority of our subjects were found to be hyperglycemic, the association of serum Na⁺ imbalance with hyperglycemia was also examined in the present study. Our results showed that there were no significant differences in the proportions of hyper-, normo-, and hyponatremia between the diabetic and nondiabetic subjects in both elderly and nonelderly groups. However, the incidence rates of hyponatremia were consistently higher in the diabetic elderly than in the nondiabetic elderly (Table VI). This result signifies that high incidence of hyponatremia at least in the elderly subjects can be explained by the increases in blood glucose levels, as well as the pathologic process of stroke discussed above.

Regarding serum K⁺ levels, hypokalemia was also more prevalent than hyperkalemia at baseline in both elderly and nonelderly subjects. This high incidence of hypokalemia appears to be associated with hyperglycemia, which commonly occurs in acute stroke patients.²² In the case of our subjects, 24% had diabetic history and approximately 90% were in a state of hyperglycemia before and after tube feeding (Table V). According to Vanlandingham et al,¹⁷ high blood glucose levels may precipitate hypokalemia by inducing insulin release, which transports potassium ions into cells. After tube feeding, the rates of hypokalemia increased in the elderly and nonelderly, whereas hyperkalemia rates were not noticeably altered. Again, high incidences of hypokalemia before and after tube feeding can also be associated with the pathologic process of stroke (ie, with hyperglycemia) and not with tube feeding.

On the other hand, whether such high incidences of hyperglycemia were only in the diabetic patients was clarified by comparing serum glucose levels between the diabetic vs nondiabetic subjects in the present study. Our results showed that no significant differences in the proportions of hyper-, normo-, and hypoglycemia were found between the diabetic and nondiabetic subjects in both elderly and nonelderly groups.

In summary, high rates of overhydration were observed in both elderly and nonelderly stroke patients before tube feeding, and this overhydration state became more prominent after iso-osmolal tube feeding in the elderly but decreased in the nonelderly. Moreover, this difference in fluid balance after tube feeding in the elderly and nonelderly was statistically significant. Contrary to fluid balance, serum Na⁺ and K⁺ concentrations were no different after tube feeding in the elderly and nonelderly subjects. Conclusively, the present study demonstrated that intake was more often positive in the elderly on days 2–3 of nasogastric tube feeding with iso-osmolal formulas but did not affect osmolality or major serum electrolyte levels, suggesting that full-strength isotonic feedings are safe and well tolerated. However, because it has been reported that elderly patients have great difficulty in maintaining a normal fluid balance due to a reduced renal concentrating ability, more attention should be paid to maintaining fluid balance in the elderly. In particular, I/O recording, body-weight measurements, and other related bedside assessments are required to detect water excesses or deficits in the elderly.

This work was supported by an Inha University research grant.

Received for publication October 6, 2005. Accepted for publication April 5, 2006.

1. Kidd D, Lawson J, Nesbitt R, MacMahon J. Aspiration in acute stroke: a clinical study with videofluoroscopy. QJM.1995; 86:825 –829.
2. Bakheit AM. Management of neurogenic dysphagia. Postgrad Med J. 2001;77:694 –699.[Free Full Text]
3. Davalos A, Ricart W, Gonzales-Huix F, Soler S, Marrugast J, Molins A. Effect of malnutrition after acute stroke on clinical outcome.Stroke. 1996;27:1028 –1032.[Abstract/Free Full Text]
4. Smithard DG, O'Neill PA, England R, et al. The natural history of dysphagia following stroke. Dysphagia.1997; 12:188 –193.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
5. Griffin KM. Swallowing training for dysphagic patients. Arch Phys Med Rehabil. 1974;55:467 –470.[Web of Science][Medline] [Order article via Infotrieve]
6. Loustau A, Lee KA. Dealing with the dangers of dysphagia.Nursing. 1985;15:47 –50.[Medline] [Order article via Infotrieve]
7. Axelsson K, Asplund K, Norberg A, Alafuzoff I. Nutritional status in patients with acute stroke. Acta Med Scand.1988; 224:217 –224.[Web of Science][Medline] [Order article via Infotrieve]
8. Aptaker RL, Roth EJ, Reichhardt G, Duerden ME, Levy CE. Serum albumin level as a predictor of geriatric stroke rehabilitation outcome.Arch Phys Med Rehabil.1994; 75:80 –84.[Web of Science][Medline] [Order article via Infotrieve]
9. Finestone HM, Greene-Finestone LS, Wilson ES, Teasell RW. Prolonged length of stay and reduced functional improvement rate in malnourished stroke rehabilitation patients. Arch Phys Med Rehabil.1996; 77:340 –345.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
10. Hoffer LJ. Clinical nutrition: protein-energy malnutrition in the inpatient. CMAJ. 2001;165:1345 –1349.[Free Full Text]
11. Suchner U, Senftleben U, Eckart T, Scholz MR, Beck K, Murr R. Enteral versus parenteral nutrition: effects on gastrointestinal function and metabolism. Nutrition.1996; 12:13 –22.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
12. Kirby DF, Delegge MH, Richard FC. American Gastroenterological Association technical review on tube feeding. Gastroenterology.1995; 108:1282 –1301.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
13. Rhoney DH, Parker Jr D, Formea CM, Yap C, Coplin WM. Tolerability of bolus versus continuous gastric feeding in brain-injured patients.Neurol Res. 2002;24:613 –620.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
14. Cataldi-Betcher E, Seltzer M, Slocum B, Jones K. Complication occurring during enteral nutrition support: a prospective study. JPEN J Parenter Enteral Nutr. 1983;7:546 –552.[Abstract/Free Full Text]
15. Gormican A, Liddy E, Thrush L. Nutritional status of patients after extended tube feeding. J Am Diet Assoc.1973; 63:247 –253.[Web of Science][Medline] [Order article via Infotrieve]
16. Panchorbo-Hidalgo PL, Garcia-Fernandez FP, Ramirez-Perez C. Complications associated with enteral nutrition by nasogastric tube in an internal medicine unit. J Clin Nurs.2001; 10:482 –490.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
17. Vanlandingham S, Simpson S, Daniel P. Metabolic abnormalities in patients supported with enteral tube feeding. JPEN J Parenter Enteral Nutr. 1981;5:322 –324.[Abstract/Free Full Text]
18. Walike J. Tube-feeding syndrome in head and neck surgery.Arch Otolaryngol. 1969;89:533 –536.[Abstract/Free Full Text]
19. Kubo W, Grant M, Walike B, et al. Fluid and electrolyte problems of tube-fed patients. Am J Nurs.1976; 76:912 –916.[CrossRef][Web of Science][Medline] [Order article via Infotrieve]
20. Miller JH, Shock NW. Age differences in renal tubular response to antidiuretic hormone. J Gerontol.1953; 8:446 –450.[Medline] [Order article via Infotrieve]
21. Bowman M, Eisenberg P, Katz B, Metheny N. Effect of tube-feeding osmolality on serum sodium levels. Crit Care Nurs.1989; 9:22 –28.
22. Mazighi M, Amarenco R. Hyperglycemia: a predictor of poor prognosis in acute stroke. Diabetes Metab.2001; 27:718 –720.[Web of Science][Medline] [Order article via Infotrieve]

Journal of Parenteral and Enteral Nutrition, Vol. 30, No. 4, 321-330 (2006)
DOI: 10.1177/0148607106030004321


CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Oh, H. Articles by Seo, W. Search for Related Content PubMed PubMed Citation Articles by Oh, H. Articles by Seo, W. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Medline Plus Health Information Fluid and Electrolyte Balance Stroke Hazardous Substances DB POTASSIUM SODIUM Social Bookmarking                   What's this?