Week 6 Assignment: Reading Research Literature (RRL) Worksheet

ARTICLE

Investigating the recording and accuracy o f flu id balance m o n ito r in g in critically ill patients A Diacon, MCur; J Bell,' 3 MCur, BCur, PGDN

1 Division o f Nursing, Faculty o f Medicine an d Health Sciences, Stellenbosch University, Cape Town, South Africa 3 TASK A pplied Science, Karl Bremer Hospital, Bellville, Cape Town, South Africa 3 D epartm ent o f Nursing Science, School o f Clinical Care Sciences, Faculty o f Health Sciences, Nelson Mandela M etropolitan University, Port Elizabeth, South Africa

Corresponding author: A Diacon ([email protected])

Background. The accurate assessment o f flu id balance data collected durin g physical assessment as well as durin g m o n ito rin g and record-keeping form s an essential part o f the baseline patien t in fo rm atio n th a t guides medical and nursing in te rven tions aimed at achieving physiological s ta b ility in patients. An inform al a u dit o f 24-hour fluid balance records in a local intensive care u n it (ICU) showed th a t seven o u t o f ten flu id balance calculations were incorrect. Objective. To id e n tify and describe current clinical nursing practice in flu id balance m o n ito rin g and measurement accuracy in ICUs, conducted as part o f a broader study in partial fu lfilm e n t o f a Master o f Nursing degree. M ethods. A q u a n tita tiv e approach u tilis in g a descriptive, e xp lo ra to ry study design was applied. An a u d it o f 103 ICU records was c o n d u cte d to establish th e cu rre n t practices and accuracy in recording o f flu id balance m o n ito rin g . Data were collected using a purpose-designed too l based on relevant literature and practice experience. Results. O f the original recorded flu id balance calculations, 79% deviated by more than 50 mL from the audited calculations. Further­ more, a significant relationship was shown between inaccurate fluid balance calculation and adm inistration o f diuretics (p=0.01). Conclusion. The m ajority o f flu id balance records were incorrectly calculated.

S AfrJCrit Care 2014;30(2):55-57. DOI:10.7196/SAJCC.193

M a in ta in in g a balance b e tw e e n flu id in ta ke and o u tp u t plays an im p o rta n t role in th e m anagem ent o f a c ritic a lly ill p a tie n t. The accurate assessment o f th e flu id balance data collected d u rin g physical assessment as w ell as du rin g m o n ito rin g activities

and record-keeping form s an essential part o f the baseline patient in fo rm a tio n th a t gu id e s m edical and nu rsin g in te rv e n tio n s to achieve physiological stability in a patient. Changes in a critically ill patient's fluid balance can com plicate the patient's clinical condition. It is, therefore, necessary th a t flu id balance parameters are accu­ rately m onitored and recorded for all patients in intensive care units (ICUs).111

A daily observation sheet is used to record all vital signs, nursing interventions, medical procedures and the fluid balance fo r each 24-h period o f a day. The fluid balance record comprises records o f the intake and ou tp u t o f fluids by a patient over a 24-h period. The difference between the volumes is calculated to provide the 24-h fluid balance.121 The m onitoring o f a patient's fluid balance is o f great im portance in understanding and managing a patient's clinical status and, as such, accurate m onitoring and recording o f fluid balance data plays an essential role in patient care management.131

Several studies have considered th e relationship between flu id imbalances and patient outcomes in critical care. The Sepsis Occurrence in Acutely III Patients (SOAP) study by Vincent et al.,m conducted across 198 ICUs in Europe in 2002, determ ined tha t a positive flu id balance is a strong prognostic factor for death in criti­ cally ill patients. Similarly, research by Alsous et a/.,151 Boyd et al.m and Payen et o/.I7] concluded th a t a more positive fluid balance

is associated w ith an increased risk o f m o rta lity in patients w ith septic shock or acute renal failure. Furthermore, Rosenberg et a/.181 determ ined tha t a cum ulative negative flu id balance in patients w ith acute lung in ju ry is associated w ith lower m ortality. The conclusions offered by these studies require th a t m on itorin g and recording o f flu id balance data must be com plete and accurate, w ith assessment o f a patient's fluid balance being recognised as an im p ortan t com ponent o f nursing any critically ill patient.

In South Africa (SA), the practice o f a registered nurse is regulated by the Scope o f Practice drawn up by the SA Nursing Council.191 Chapter 2, section 2(i) o f these regulations identifies that fluid balance m onitoring is part o f the scope o f practice o f a registered nurse. Therefore, a registered nurse working in a critical care environment is responsible and accountable fo r the accurate recording and calculation o f fluid balance when caring fo r and managing a critically ill patient. Managing a patient's fluid balance is as equally im portant as carrying out any other patient care activity for the critically ill, such as administering a medication prescription or providing nutrition.121

Fluid balance management in ICU patients is complex. M onitoring and measurement o f fluid balance requires close attention to ensure that current methods are applied accurately and consistently to provide the most complete data, upon which patient management decisions can be based.

Based on practice experience and underpinned by an informal audit o f 24-h fluid balance charts in a local ICU, where seven out o f ten calculated totals were incorrect, the research question posed was: What are the current practices o f registered nurses in ICUs w ith regard to fluid balance monitoring?

SAJCC November 20 14, Vol. 30, No. 2 55

Methods A q u a n tita tiv e approach u tilis in g an exploratory, descriptive study design was applied. The study was conducted in ICUs across three purposively selected hospitals o f one private sector hospital group. The ICUs o f these hospitals were similar in terms o f their patient admission profiles, with the same nursing documentation and policies applied at all three hospitals.

An audit tool was developed from relevant literature and clinical experience to assess particular aspects of the sampled fluid balance records. Two critical care nurse experts evaluated the content and face validity of the audit tool; no changes were required. A pretest of the audit tool was conducted at one additional ICU of the same hospital group to determine the accuracy and relevance of the measurements; no changes were required. The pretest data were not included in the study data. A statistician determined the tool to be appropriate and adequate for data collection and analysis purposes.

Ethical approval for the study was obtained from the Human Research Ethics Committee at the Faculty of Medicine and Health Sciences, Stellenbosch University, as well as the relevant committee of the hospital group.

The population for this study was critical care patient records. The study sample was drawn from fluid balance records according to the following inclusion criteria: • Nursing records of admissions to ICUs for

the first 48 h of the patient's stay, from 1 July to 31 December 2011

• Patients over the age of 18 years as per the definition of an adult in the Children's Act No. 38 of 20051'01

• Patients classified as 'intensive care': activity 1 or 2 on the patient classification system o f th is h o sp ita l group. This classification was used by the doctor

to determ ine financial charges to the patie n t. No w ritte n p o licy regarding this classification was available from the hospitals.

A simple random sampling technique was implemented to select patient records for the audit: all the admission numbers of patients meeting the inclusion criteria were identified through the hospital inform a­ tio n system and admission record book of the ICU. The p atient record file th a t was connected w ith every th ird patient admission num ber was drawn until the required sample was achieved. The sample size was calculated to ensure adequate precision in population estimates, using 95% confidence intervals (CIs). A sample size of 80 fluid balance records would have resulted in 6% precision in the 95% Cl width, assuming a 10% error rate in the calculation ofthefluid balance. This was well w ithin the accepted precision of between 5% and 10%. A sample size of N= 103 was selected and divided specifically among the various units under the guidance of the statistician (Table 1). Descriptive statistics were recorded and the Mann-Whitney U-test was used to test associations between recorded variables and fluid balance calculation accuracy.

Data were recorded on the study audit tool by the researcher and a field worker together in the three hospitals. The fluid balance calculation recorded in each

patient record for a 24-h period during the first 48 h of a patient's stay was noted on the audit tool. A control calculation of each recorded fluid balance total was done by the researcher and verified by the field worker. These audited calculations were recorded in the audit tool. The deviation between the original calculations and the audited calculations was determined and recorded.

In addition to the flu id balance calculation, baseline vital sign data, modes o f fluid output (e.g. diarrhoea), specific data regarding the administration of blood products, and the number of continuous intravenous infusions were recorded on the audit tool.

Results 24-h calculated fluid balance totals The original recorded 24-h fluid balance total was compared with the audited fluid balance total performed by the researcher and field worker. The difference in calcula­ tion was referred to as the deviation in fluid balance calculation, and is presented in Table 2 for descriptive reasons.

In the audit of 103 fluid balance documents, a total of 71 (68.9%) recorded calculated fluid balance totals were within a 500 mL deviation from the fluid balance calculated by the researcher. Fourteen recorded calculations (13.5%) were found

T a b le 1. S a m p lin g fra m e w o rk

Hospital Intensive-care beds, n Admissions: July - D ecem ber 2 0 1 1 , n Records sampled, n

A 26 1 020 34

B 28 1 027 34

C 38 1 022 35

D 12 300 Pilot study

Table 2. Deviation in fluid balance (A/=103)

Calculated d eviation Overall 0 - 3 706 0 - 50 51 - 500 501 - 1 000 1 001 - 2 000 >2001 No record

n 98 22 49 14 7 6 5

Percentage 95.1 21 48 13.5 6.8 5.8 4.9

Median deviation (mL) 167 20 146 754 1 249 3 310 -

Mean deviation (mL) 493 21 184 754 1 371 3 116 -

Range (mL) 0 - 3 706 0- 4 6 61 -463 501 - 984 1 008- 1 928 2 260 - 3 706 -

56 SAJCC November 2014, Vol. 30, No. 2

Table 3. Comparison of accurate and inaccurate fluid calculation

Inaccurate flu id calculation, m edian (IQR)

V a ria b le Yes No p -v a lu e

Received blood products 180.5 (60- 1 312) 167 (61 -530) 0.95 CVP measured 202.5 (90 - 764) 119 (41 -320) 0.09 Matched doctor's prescription 155 (60 - 530) 201 (63 - 708) 0.61

Diuretic administered 279(102-996) 106 (46 - 350) 0.01

Received >2 intravenous drugs 257 (75 -708) 138 (60-435) 0.16 IQR = in te rq u a rtile range; CVP = central venous pressure.

to deviate between 500 mL and 1 000 mL, while seven recorded calculations (6.8%) were found to deviate between 1 000 mL and 2 000 mL. Six recorded calculations (5.8%) were found to have a deviation of >2 000 mL.

There was a significant association between the administration o f diuretics and inaccurate fluid balance calculation (p=0.01), but there was no association between other variables and the outcome o f interest (Table 3).

Discussion The definition of a net positive fluid balance as a volume >500 mL used in the study by Alsous eta/.151 was applied in this study. Of great concern were the 27/103 documents, more than 25% o f the sample, w ith a deviation of >500 mL between the recorded calculation and the control calculation. Equally o f concern were the five patient records where no fluid balance calculation was available at all. These findings repres­ ent a risk for the critically ill patient when one considers the findings o f previous studies related to positive fluid balance and patient m ortality.14'81 The findings o f this study showed that fluid balance calculation is not treated as a priority in the nursing m anagem ent o f a c ritic a lly ill p atient. The incorrect calculation o f fluid balance means th a t every p atient management decision utilising these fluid balance data was influenced by inaccurate information. Perren et al.1" 1 performed a similar study in Switzerland and expressed th e ir concern about the accuracy o f flu id balances in critically ill patients.1111

Additionally, the significant association between inaccurate fluid balance calcula­ tion and diuretic administration (p=0.01) suggests that when diuretics are adminis­ tered, there is a higher chance of the

calculated fluid balance being incorrect.This finding supports the researcher's concern that a careful and accurate approach to fluid balance does not enjoy high priority in managing critically ill patients in this context. Diuretic therapy is a commonly prescribed therapeutic modality; in this study, 38.8% (40/103) of critically ill patients had diuretics recorded as being adminis­ tered during the first 48 h of their admission. Inaccurate fluid balance data may result in inappropriate application of diuretic therapy, resulting in fluid imbalances that affect the haemodynamic stability of patients.

The findings of this study are limited by the focus on one hospital group and may be regarded as a pilot study for further development.

Conclusions in this study, the majority o f audited 24-h fluid balance calculations were shown to be incorrect; 79% (81/103) o f the original recorded fluid balance calculations deviated by >50 mL from the audited calculation. The accuracy o f the 24-h balance calculated is questionable, with only 21% of the original fluid balance totals deviating by <50 mL from the audit calculations. This is of great concern. Several studies14'81 have noted a relationship between flu id imbalance and m ortality in critically ill patients. The findings indicate that treatment decisions are often based on inaccurate fluid balance inform ation, which may lead to negative consequences for the patient.

A significant association was shown between the administration of diuretics and inaccurate 24-h fluid balance calculations. With diuretics prescribed specifically to manage fluid imbalance, this finding indicates that the accuracy of the calculated fluid balance must be confirmed prior to diuretics being prescribed or administered.

Within the context of limited resources, any clinical recommendations must be realistic and practical. One suggested example is in stitu tin g a system o f checking fluid balance calculations at specific intervals, such as during patient handover at shift change, during the patient assessment process or during p atient management discussions. Awareness around the poten­ tial consequences of calculation errors must be reinforced during patient discussions and continuing education sessions.

The requirement to provide accurate, correct fluid balance monitoring and recording as part of the patient's vital sign data must be established as a fundamental standard of practice for every nurse practising in an ICU. Regular outcome- driven audits will assist in identifying where and when errors occur, allowing for specific interventions to be designed and implemented.

Further studies may assist in refining the particular challenges of accurate fluid balance recording, for instance cumulative fluid balance over more than 24 h.

References 1. Culleiton AL, Simko LC. Keeping electrolytes and fluids in

balance. Nursing2013 Critical Care 2011;6(2):30-35. [http:// dx.doi.org/10.1097/01 :CCN.0000394395.67904.4d]

2. Scales K, Pilsworth J. The importance of fluid balance in clinical practice. Nursing Standard 2009;22(47):50-57. [http://dx.doi.org/10.7748/ns2008.07.22.47.50.c6634]

3. Elliot D, Aitken L, Chaboyer W. ACCCN's Critical Care Nursing, 1st ed. Marrickville, Australia: Mosby Elsevier, 2007:440,445-446.

4. Vincent JL, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: Results o f the SOAP study. Crit Care Med 2006;34(2):344-353.

5. Alsous F, Khamiees M, DeGirolamo A, Amoateng-Adjepong Y, Manthous CA. Negative fluid balance predicts survival in patients w ith septic shock: A retrospective study. Chest 2000;117(6):1749-1754.

6. Boyd JH, Forbes J, Nakada T, Walley K, Russell JA. Fluid resuscitation in septic shock: A positive fluid balance and elevated central venous pressure are associated with increased mortality. Crit Care Med 2011;39(2):259-265. [http://dx.doi.Org/10.1097/CCM.0b013e3181 feebl 5]

7. Payen D, de Pont AC, Sakr Y, et al. A positive fluid balance is associated with a worse outcome in patients w ith acute renal failure. Crit Care 2008;12(3):R74. http://ccforum.com/ content/12/3/R74 (accessed 27 October 2014). [h ttp:// dx.doi.org/10.1186/cc6916]

8. Rosenberg AL, Dechert RE, Park PK, Bartlett RH, National Institutes of Health-National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network. A review of a large clinical series: Association of cumulative fluid balance on outcome in acute lung injury: A retrospective review o f the ARDSnet tidal volume study cohort. J Intensive Care Med 2009;24(1):35-46. [h ttp:// dx.doi.org/10.1177/0885066608329850]

9. South African Nursing Council. R2598, Regulations relating to the scope of practice o f persons who are registered or enrolled under the Nursing Act, 1978. Regulation of the Nursing Act, 2005 (Act No. 33 o f 2005). Pretoria: Government Gazette, 491,2006:34.

10. South African Government. Children's Act 38 o f 2005. http://www.justice.gov.za/legislation/acts/2005-038%20 childrensact.pdf (accessed 27 October 2014).

11. Perren A, Markmann M, Merlani G, Marone C, Merlani P. Fluid balance in critically ill patients. Should we really rely on it? Minerva Anestesiol 2011;77(8):802-811.

SAJCC November 2 0 1 4 , Vol. 30, No. 2 57

Copyright of Southern African Journal of Critical Care is the property of Health & Medical Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.