The Risk Of Cerebral Oedema In Children Nursing Essay

ABSTRACT

OBJECTIVES

Over-hydration in diabetic ketoacidosis (DKA)

may increase the risk of cerebral edema in

children.

Patients

Fourty five pediatric patients (1 month–16 years) presenting with 41 episodes of DKA.

Intervention The clinical and biochemical variations was selected by the hospital during the date of their hospital admission. Dehydration was calculated by measuring acute changes in body weight during the period of illness.

Results

The magnitude presentation of the centile 25~35 percent was in the ratio of 5.6% (3.4–8.2%) (mean±SD 6.1±4so there was no clinical and biochemical assessment variation was needed. These both of the variations ,all of the diabetic ketoacidosis patient approached. Further all the patient variation w not correlated with the amplitude of variation and magnitude presentation and did not affect the fluid concentration and the quantity of the fluid was 47.8 ml/kg (36.5–56.3)) in the first 12 hours.

Conclusion

For the conclusion of the exact perimeters and the magnitude variations of the fluid in the patients of diabetic ketoacidosis. All of the conformations needs study on the larger scale.

MATERIALS AND METHODS

We conducted a prospective study of consecutive patients aged 1 month–16 years, diagnosed with DKA, who presented to the emergency department in a xinjiang medical university hospital

from January 2010 to February 2012.The study was approved by the Ethics Committee of the First Affiliated Hospital of Xinjiang Medical University, this study was done in the declaration of Helsinki.

. Patients were excluded from the study when patient and/or parental consent was not obtained, or when fluid replacement therapy had begun prior to weighing the patient. DKA was defined as a pH <7.3 (venous), a glucose >11 mmol/l, calculated

bicarbonate (HCO3) of <15 mmol/l and the presence of urine ketones.1 Recovery from DKA was defi ned as HCO3 ≥18 mmol/l. Blood gases were assessed at least every 2 h in the first

12 h and at least every 4 h in the next 24 h. The following clinical variables were collected on admission: demographics, heart rate, arterial blood pressure, capillary refill time and level of consciousness using the Glasgow Coma Scale (GCS). The laboratory results recorded on admission included: blood gases, glucose, hemoglobin, haematocrit, electrolytes, osmolality, creatinine, blood urea nitrogen (BUN), HbA1c, albumin and lactate levels. The osmolar gap (measured osmolality−calculated osmolality) was calculated. 12 Sodium (Na+) was corrected for glucose level ([Na+]+([ glucose]−5.6)/5.6)×1.6). Deviation of the heart rate and blood pressure from the mean and maximum normal values for age were calculated for each patient.12 Weight was obtained on admission before fluid administration (admission weight).

and every 12 h thereafter until discharge from hospital (discharge weight) and until two consecutive similar weights (0.5%) were obtained (final weight). In the absence of two

consecutive similar weights, the patient’s weight within 7 days after discharge was considered as the final weight. The magnitude of dehydration was defined as the % loss of

body weight (LBW) ((discharge weight−admission weight)/ discharge weight×100).. An average of three readings were recorded. The calculated coefficient of variation (SD/

mean×100) of the scales was a mean of 0.21% for patients over 15 kg and 0.34% for those under 12 kg. After the study, we retrospectively examined the charts for the actual amount of fluid administered to each patient. For determining the percentage of fluid above maintenance, the

maintenance amount of fluid was calculated according to the following: 100 ml/kg for the first 10 kg, 50 ml/kg for the next 10–20 kg and 20 ml/kg thereafter12 All patients were managed according to a pre-existing provincial management protocol (endorsed by the Province of china xinjiang) for fluid, insulin and electrolyte administration In brief, on admission, each patient initially received 10 ml/kg measured weight of 0.9% saline, repeated as necessary until haemodynamic stability was achieved as judged by the responsible physician. This was followed by an infusion o f 0.9% or 0.45% saline at a rate of 4–6 ml/kg/h (independent of age and urine output) until DKA resolved. Potassium and

glucose were added to the solution as necessary.

Table 1 Demographic characteristics of all analyzed patients on admission with diabetic ketoacidosis episodes

Age (years)

11 (5–16)

New diabetes (%)

52.6

Male (%)

47.8

Weight (kg)

35.4 (18.3–56.9)

Fluid (ml/kg/12 h)*

52.9 (45.0–66.7)

Fluid over calculated maintenance (%)

107 (51–148)

Calculated degree of dehydration (%)

5.7 (3.7–8.1)

Length of stay (h)

30 (24–48)

Duration of symptoms (days)

5.0 (2–16)

PCCU admission (%)

21.1

pH

7.12 (7.05–7.22)

Na+ (mmol/l)

133 (130–135)

Na+ corrected (mmol/l)

140.4 (136.9–143.9)

Glucose (mmol/l)

33.3 (25.2–43.6)

Chloride (mmol/l)

96.5 (94–103)

Haematocrit 0.44 (0.41–0.48)

0.44 (0.40–0.48)

Hemoglobin (mg/dl)

13.3 (12.3–14.7)

*First 12 h including resuscitation fluid. PCCU, pediatric

Statistical analysis

SPSS V.16.0 was used for data analysis. Categorical data were summarized as percentages. The results were reported as medians (25th–75th centiles).

The Pearson correlation coefficient was used to assess the relationship between normally distributed continuous variables, and Spearman.s rho was used for skewed continuous

variables. The χ2 test or Fisher’s exact test was used to compare differences in proportions for categorical variables. The independent samples t test was used to compare mean differences

in normally distributed continuous outcomes, and the Mann–Whitney U test was used to compare skewed continuous variables between categorical independent variables. A p value of <0.05 was considered statistically significant except when the Bonferroni adjustment was made for multiple comparisons.

RESULTS

During the study period, 49 patients with 53 episodes of DKA were managed in our centre. Six patients had commenced treatment prior to arrival and were not eligible for the study. Four

of the 43 eligible patients declined to participate. Therefore, 39 patients with 42 episodes of DKA were successfully enrolled and completed the study.

Demographics

Demographic data are provided in table 1. All patients had type 1 diabetes. Nine of the patients (21%) were admitted to the pediatric intensive care unit. The rest were admitted to

the pediatric medicine department. Only three patients (7.35 %) presented with a GCS <13. One patient (2.69%) presented with mild hypothermia (32.9°C) and required mechanical ventilation.

No patient experienced a decreased GCS after initiation of therapy and all had attained a GCS >13 within 12 h of commencing treatment

TABLE 2

Association between demographic, clinical and biochemical variables recorded on admission and magnitude of dehydration

Categorical variables p* Value

PCCU admission

0.783

New diabetes

0.623

Lactate >2.2 mmol/l

0.748

Continuous variables

r Value

p‡ Value

Age (years)

−0.192

0.216

pH 0.146 0.363

0.1444

0.360

HCO3 0.057 0.720

0.055

0.718

Glucose

0.203

0.200

Sodium [Na+]

0.355

0.018

Corrected sodium [Na+]

0.501

0.001

Potassium (K+)

0.210

0.172

Creatinine

0.207

0.179

Blood urea nitrogen

0.276

0.071

CO2

0.040

0.790

Creatinine

0.209

0.179

Creatinine†

0.281

0.065

Anion gap

0.047

0.632

Anion gap (corrected)

0.050

0.752

HbA1c

−0.028

0.870

Osmolality

0.223

0.151

Osmolality (gradient)

0.003

0.984

Fluid (ml/kg/12 h)

0.111

0.477

Fluid (% above maintenance)

0.011

0.935

Mean heart rate†

0.041

0.777

Respiratory rate

− 0.118

0.448

Mean blood pressure†

− 0.101

0.427

Length of stay

0.097

0.557

Duration of symptoms

−0.154

0.307

*Calculated using the t test and Mann–Whitney U test.

†Levels above normal values corrected for age.

‡Calculated using the Pearson correlation coefficient for normally distributed

continuous variables; Spearman’s rho was used for skewed continuous variables.

PCCU, pediatric critical care unit

Correlation of dehydration with biochemical or clinical

parameters on admission

In the study population, the median (25th–75th centiles) magnitude of dehydration, based on LBW, was 5.7% (3.7–8.3%) (range 1.4–24.8%) or 8.9% (6.0–13.5%) (range 1.9–41.0%) when corrected for % of total body water. In 20/42 (48%) episodes, parents had noted weight loss in their child. No

significant correlations were found between the patient’s demographic variables (age, duration of symptoms, new versus known diabetes), clinical variables (deviation of heart rate

from maximal rate corrected for age or deviation of mean arterial blood pressure corrected for age) or biochemical variables (pH, HCO3, glucose, CO2, HbA1c, creatinine and BUN deviation

from maximum normal values, anion gap, corrected anion gap, osmolar gap and the presence of lactic acidosis) recorded on admission and the magnitude of dehydration, whether

based on LBW or corrected for % of total body water (table 2). Sodium and corrected sodium level were the only biochemical variables to show reasonable correlation with severity of

dehydration (r=0.35, p=0.02 and r=0.502, p=0.001, respectively). Although no significant correlation between patient age and the magnitude of dehydration was observed across the whole group, when we subdivided the patients, based on age, into those ≤2 years and >2 years of age, the magnitude of dehydration in children ≤2 years of age (n=6) was significantly greater whether expressed in LBW (9.7% (6.2–11.8%) vs 5.3% (3.5–7.4%), p=0.014) or corrected for total body water (15.2% (10.5–18.5%) vs 8.3% (5.5–10.7%), p=0.016).

Categorizing the magnitude of dehydration to non-severe

and severe

Severe dehydration (LBW ≥6% in children, ≥10% in infants)12 was detected in 18/42 (43%) patients. The main characteristics of patients with non-severe (mild, moderate) and severe

dehydration are presented in table 3. Except for the corrected [Na+] on admission, no significant differences were observed between groups.

Correlation between magnitude of dehydration and amount

of fluid administered

In the first 12 h, fluids were administered exclusively via the intravenous route. The median (25th-75th centiles) volume of administered during the first 12 h was 52.9 ml/kg (45-66.4 ml/kg). No correlation was detected between the total amount of fluid administered or the amount of fluid administered

above maintenance and the magnitude of dehydration (figure 1). There was, however, a negative correlation between the amount of fluid administered during therapy and the pH on admission (r=−0.44, p<0.05).

Resolution of DKA

In all patients, DKA resolved within 24 h (figure 2). No significant elations were found between the patient’s age, ew versus known diabetes, glucose level, osmolar gap, presence

of lactic acidosis, magnitude of dehydration on admission r amount of fluid administered in the first 12 h (whether expressed as ml/kg or % above maintenance) and the rapidity

of recovery from DKA (figure 2). However, there was a significant negative correlation between the level of HCO3 on admission and time to recovery (r=−0.74, p<0.001).

DISCUSSION

Fluid replacement is the cornerstone of therapy in DKA. Work n animal models demonstrates that successful fluid replenishment lone will reverse many of the clinical and biochemical

derangements seen in DKA. However, precise fluid therapy is often impeded by difficulties in the determination f the magnitude of dehydration in DKA and differences in

opinion regarding the need to account for on-going fluid (urinary) loss. The mean magnitude of dehydration observed in his study is similar to that reported in previous studies4 and

supports current recommendations to assume a 7–10% depletion n total body water when managing patients in DKA.1 14 these recommendations are based mainly on two studies

performed in 195215 and 193316 in a small number of patients. he observation that children ≤2 years of age had a significantly higher degree Of dehydration has also been reported

previously.17 It is speculated that younger children, who can either independently gain access to fl uids nor successfully ommunicate symptoms, experience a greater degree of

dehydration and acidosis.18 19 despite reports that severe dehydration is rare in

DKA,1 4 5 severe dehydration was observed in 18/42 (43%)Patients in this study, a higher proportion than recorded in he studies of Harris et al 7 (23.7%) and Fagan et al 4 (12%).

This difference could be due to different patient demographics, Such as the large geographical size of our referral region and subsequent delays in transport for definative treatment.16

Table 3

Comparison of vital signs, serum markers and demographics between those with mild, moderate

and severe dehydration*

Non-severe dehydration (n=24) Severe dehydration (n=18) p Value

Age (years)

11.5 (7–15)

11.0 (5–14.2)

0.703

Duration of symptoms (days

6 (3–28)

5.0 (1–14)

0.40

Length of stay (h)

30 (24–48)

30 (24–49

0.858

PCCU admission (%

5 (19.8)

4 (21.2)

0.74

New diabetes (%)

13 (54.6)

8 (49.0)

0.98

Delta heart rate mean (bpm

42.7 (12.4–52.8)

41.9 (30.0–51.6)

0.70

HbA1c (%)

11.9 (10.8–12.9)

11.2 (10.0–13.6)

0.35

pH

7.11 (7.03–7.22)

7.20 (7.05–7.21)

0.35

HCO3 (mmol/l)

8 (6–12.5)

10.5 (6–11)

0.46

Anion gap corrected

27.1 (25.4–33.2)

28.6 (25.3–29.9)

0.80

Glucose (mmol/l)

32.1 (25.8–42.9)

35.7 (27.9–43.9)

0.32

Blood urea nitrogen (mmol/l)

6.9 (4.6–9.6)

7.5 (6.5–12.2)

0.19

Cr (μmol/l)

90 (61.5–133.5)

105 (75–136)

0.559

Lactate (mmol/l)

2.2 (1.4–2.6)

2.3 (1.5–3.3)

0.51

Na+ (mmol/l)

133 (128–135)

134 (132–136)

0.273

Na+ corrected (mmol/l)

138.2 (136.5–143.0)

141.3 (140.7–144.5)

0.036

Osmolality (mOsm/kg)

329 (317–337)

329 (320–350

0.413

Osmolar gap (m0sm/kg)

46.4 (41.5–58.7)

47.9 (37.2–62.2)

0.95

Fluid (ml/kg/12 h)

50.2 (34.7–58.1)

51.6 (45.8–59.4

0.89

Fluid over the calculated maintenance (%)

111 (60–145)

91 (66–146)

0.60

Values are median (25th–75th centiles).

*Dehydration was subcategorized based on change in body weight during diabetic ketoacidosis (severe: >6% in children,

>10% in infants).

AG, anion gap; PCCU, paediatric critical care unit.

This study did not demonstrate a significant correlation between comprehensive biochemical, clinical or demographic characteristics of the patients and magnitude of dehydration.

This is consistent with recent studies, which indicate that clinicians failed to predict the magnitude of dehydration accurately when employing more selective clinical or biochemical

patient data.4 5 8 20 It is therefore not surprising that the amount of fluid administered did not correlate with the magnitude of dehydration (figure 1). We speculate that the replacement fluid administered reflects the physician’s subjective impression of the patient’s degree of distress. However, the patient’s ‘ill’ appearance in DKA may be affected by factors other than fluid deficit, such as acidosis, which can alter level of consciousness, affect perfusion by vasoconstriction and stimulate Kussmaul’s respirations which results in dry mouth.5 It is these factors that may explain the negative correlation between pH and the observed amount of fluid administered in this study. The inability to clinically estimate the level of dehydration in children with DKA with accuracy may in part reflect a relatively well preserved intravascular volume. Furthermore, urine output(usually a sensitive marker of dehydration) is maintained

by a glucose driven diuresis.21All patients were treated in accordance with a provincial

guideline for DKA management, consistent with the ESPE/ LWPES consensus statement. The median volume of fluid administered in the first 12 h of therapy was 3.7 ml/kg/h (3.0–4.6%) (range 2.1–8.3ml/kg/h), which is a more restrictive (about 25% less) regimen than recommended in the literature.22 Despite a wide range of fluid deficit observed in this study, all patients achieved resolution of DKA within 24 h of treatment initiation, independent of severity of dehydration or amount of fluid administered (figure 2). The patients’ weight gain achieved between hospital admission and DKA resolution (median 3.4%, data not presented) was approximately 65% of the patient’s calculated dehydration (median 5.6%). Hence,

Figure 1 Correlation between fluid administered and magnitude of

dehydration, including initial boluses given on admission (r=0.23,

p>0.05).

this suggests that biochemical resolution in DKA is achieved without full rehydration of the patient. In the absence of a validated method of assessing dehydration in DKA, one should be careful about using the patient’s clinical status to drive fluid replacement therapy. No patient experienced permanent neurological impairment. This suggests that, unless the patient is haemodynamically unstable, it is not necessary to aggressively correct fluid deficit. We believe this is an important and novel finding, as one can then assume that the restrictive strategy of ‘one size fits all’ for rehydration is safe and effective. The validity of this study is subject to a number of limitations. Caution needs to be taken when interpreting several of the study findings because our patient group is relatively small.

Figure 2 Relationship between time to diabetic ketoacidosis

resolution and degree of dehydration (column), volume of fluid

administered in the first 12 h of therapy (■) and the admission serum

bicarbonate level (Δ). Significant correlation (negative) was found only

between the level of bicarbonate on admission and time to resolution

(r=−0.74, p<0.001).

and the groups are unequal. For example, the findings in table 3 that do not show significant differences may be a reflection of lack of power. Given our small sample size, as well as

the weak correlations and non-significant relationships of the majority of the biochemical and clinical variables with severity of dehydration (table 2), a multivariable regression model

was not performed, which would have allowed us to control for confounding variables (table 2). Although acute weight loss is accepted as the gold standard for quantifying the magnitude

of dehydration,4 6 7 the robustness of the correlation is somewhat understudied. Furthermore, some may argue that the final weight of the patient better reflects total fluid loss rather

than the discharge weight which was used in our study and by others,4 5 especially since DKA and, in particular, new onset DKA, is characterized by an acute catabolic state, which results

in weight reduction from the loss of lean body mass, as well as from concomitant dehydration. We believe that the final weight reflects the weight gain of lean body mass, which occurs with the introduction of insulin, rather than fluid loss alone.

CONCLUSION

Children in DKA present with a median (25th–75th centiles)dehydration of 5.7% (3.4–8.3%) (mean}SD 6.8}5.0%); however, there is large variability among patients (range 1.4–24.8%).

Since the magnitude of dehydration cannot be assessed accurately by either clinical or biochemical parameters, clinicians should not attempt to administer fluid according to subjective estimates of dehydration. Unless the patient is haemodynamically unstable, a relatively restrictive regimen of fluid administration

is associated with timely correction of DKA