Secondary Logo

Journal Logo

Clinical Science Aspects

Shock in the First 24 h of Intensive Care Unit Stay

Observational Study of Protocol-Based Fluid Management

See, Kay Choong*†; Mukhopadhyay, Amartya*†; Lau, Samuel Chuan-Xian; Tan, Sandra Ming-Yien; Lim, Tow Keang*†; Phua, Jason*†

Author Information
doi: 10.1097/SHK.0000000000000332
  • Free



Fluid management for shock is a cornerstone of intensive care unit (ICU) practice, particularly for severe sepsis (1–3). Guidelines advocate large-volume fluid loading within the first few hours of shock (4). However, it is often difficult to differentiate between septic and nonseptic conditions, to diagnose the precise causes of shock, and to predict the response of patients to fluid loading (5–7).

Variable methods of fluid challenge and poor documentation of volume response end points could potentially lead to either under-resuscitation or unnecessary repeat fluid boluses (8, 9). Despite data showing harmful consequences of excessive volume resuscitation, little information exists to show that titrating fluid therapy precisely can improve clinical outcomes (10–13). Also, no method of tailoring fluid administration has been shown to be superior to another.

We frequently encountered nonuniform fluid resuscitation practices and inadequate documentation for patients who developed shock in our ICU. As part of a quality improvement effort, we thus introduced protocol-based fluid management (PBFM) in August 2011 to facilitate timely and appropriate fluid administration. Our protocol was meant to expedite fluid resuscitation in a careful and pragmatic fashion, and we did not use central venous oxygen saturation or central venous pressure monitoring (14). Although the PBFM became the standard of care in our ICU, some physicians remained confident with using their own clinical assessment and management for some patients. We postulated that adherence to PBFM would be linked to better clinical outcomes. We thus audited protocol adherence when patients developed shock within 24 h of ICU stay and investigated associations of adherence with ICU mortality, hospital mortality, ventilator days, ICU length of stay (LOS), and hospital LOS.


Study design

We conducted an observational study in the 20-bed medical ICU of our 1,081-bed university hospital. Because the fluid management protocol was introduced as standard care and because the study was performed as part of an ICU audit, our ethics review board permitted a waiver of informed consent (DSRB B/2013/00132).


We included all mechanically ventilated patients admitted directly from the emergency department (ED) to our ICU from August 2011 to December 2013, who also developed shock within the first 24 h of ICU stay. We chose to include only mechanically ventilated patients because these patients routinely received arterial cannulation and invasive blood pressure monitoring. Only the first ED to ICU admission was used in patients with multiple encounters during the study period. Patients with onset of shock beyond 24 h of ICU stay were excluded.

The presence of shock was defined as systolic blood pressure less than 90 mmHg or lactate greater than 4 mmol/L within the first 24 h of ICU stay. All systolic blood pressure readings less than 90 mmHg were routinely rechecked and confirmed by nurses before documentation. We did not consider a drop of more than 40 mmHg from baseline blood pressure because the latter was largely unknown in our practice. Shock Index was computed as heart rate divided by systolic blood pressure (15). Sepsis was defined according to the 1992 American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference criteria, which required the presence of the systemic inflammatory response syndrome caused by infection (16).

General clinical management

In the absence of PBFM, physicians used their clinical judgment to prescribe crystalloid fluids to patients who they felt would benefit. Volume, rate, process, and hemodynamic end points of fluid administration varied among physicians. Arterial line placement was usually in the radial position and, alternatively, in the femoral position. Use of a cardiac output monitor to obtain stroke volume (SV), the Flotrac-Vigileo device (software version 1.10; Edwards Lifesciences, Irvine, Calif), was optional (17). While on controlled mechanical ventilation, tidal volumes were kept at 6 mL/kg ideal body weight as far as possible. Nurses administered sedation to achieve a Richmond Agitation Sedation Scale score of -2 to 0 and avoid neuromuscular paralysis. Respiratory therapists awakened and assessed patients for extubation daily. Patients diagnosed as having sepsis had early blood and urine cultures and received empiric antibiotics.

Protocol-based fluid management

Because no universally recognized method existed, the protocol steps were designed after a consensus discussion among our 15 ICU physicians (18, 19). Important elements were close coordination between physicians and bedside nurses, prompt documentation, and computerized calculation assistance. To promote timely fluid resuscitation, we simplified protocol entry, such that its use was not dependent on the etiology or mechanism of shock.

A physician order was required to execute the protocol. Thereafter, the bedside nurse carefully instituted the fluid challenge in three steps, providing a total of 500 mL of Ringer’s lactate solution within 25 min (Table 1). We chose Ringer’s lactate solution because it was the least costly balanced crystalloid available in our hospital formulary. Assuming no major changes in central arterial impedance, pulse pressure ([PP] difference between systolic and diastolic blood pressures) would be a reasonable surrogate for SV (19). We chose to reassess patients after every 250 mL infused rather than after a single 500-mL bolus, which meant that we could do two fluid challenges with the same total fluid volume and could improve sensitivity of arterial PP monitoring (18–20). Flow rate was fixed at a maximum of 1,200 mL/h using infusion pumps (Colleague Volumetric Infusion Pump; Baxter Healthcare, Singapore). All blood pressure readings were immediately and directly entered into the computerized ICU flowsheet, which autocalculated the PP, the change in PP between Step 1 and Step 2 (PP1-2), and the change in PP between Step 2 and Step 3 (PP2-3). When available, the change in SV between Step 1 and Step 2 (SV1-2) and the change in SV between Step 2 and Step 3 (SV2-3) were also computed. Notably, we used PP and SV changes as our fluid response end points. We avoided using dynamic PP and SV variations, which predict volume responsiveness, because these are limited by the need for deep sedation, regular cardiac rhythms, and ventilator tidal volumes larger than those recommended for lung protection (8, 21). Vasoactive medications could not be introduced or changed during the protocol-based fluid challenge.

Table 1:
Fluid challenge protocol

On completion of the fluid challenge, ICU physicians were then called to review the patient immediately. A change of 10% or more in PP1-2, PP2-3, SV1-2, or SV2-3 was taken as a positive response to fluid challenge and triggered successive 500-mL crystalloid boluses until the PP or SV change was less than 10% or until the hypotension resolved (6). The results of these fluid boluses were again promptly recorded in the flowsheet. A 10% threshold was maintained regardless of cardiac rhythm to simplify the protocol and ease implementation, although this would likely lead to decreased specificity for volume response in patients with atrial fibrillation (22). Negative responses to fluid challenge or fluid boluses triggered a consideration for vasopressor use. The protocol could be repeated for persistent or recurrent shock.

Data collection

K.C.S., S.C.L., and S.M.T. retrospectively extracted data from medical records. Because all PBFM steps were recorded in the ICU computer, adherence could be accurately determined. Diagnosis was taken as the main diagnosis recorded in the ICU admission notes. The predominant cause of shock was derived from documented clinical judgment at the point of ICU admission. Glasgow Coma Scale scores were taken as the worst score before any sedation or intubation. Antibiotic therapy was considered discordant if blood or urine cultures were resistant to all of the empiric choices.

Statistical analysis

Univariate comparisons of proportions, means, and medians were respectively done using Fisher exact (or chi-square), Student t, and Wilcoxon rank-sum tests. Logistic regression models for mortality included covariates with P < 0.1 on univariate analysis. Year of ICU admission was included as a categorical variable to evaluate any association of mortality across time.

To understand how PP and SV changes contributed to the determination of fluid challenge results, we collected more detailed data for the fluid challenges done using PBFM for patients in 2012, only if the arterial line was connected to the Flotrac-Vigileo device. These fluid challenges could be done at any time, including beyond the first 24 h of ICU stay. We assessed the sensitivity, specificity, percent agreement, and kappa of positive PP1-2 (versus positive SV1-2 as a reference standard), positive PP2-3 versus positive SV2-3, and positive PP1-2 or PP2-3 versus positive SV1-2 or SV2-3 (23). Finally, using overall protocol-defined volume response as the reference, we recomputed the test characteristics of consecutive 250-mL fluid challenge boluses and 500-mL fluid challenges for both PP and SV end points.

Statistical significance was taken as P < 0.05. Parts of this study were submitted as abstracts to the Singapore Health and Biomedical Congress 2014 and the Asian Pacific Society of Respirology Congress 2014.


During the study period, we had 1,439 ICU admissions and 757 mechanically ventilated patients. Within the first 24 h of ICU stay, 612 mechanically ventilated patients (mean [±SD] age, 63.0 years [16.5]; 252 or 41.2% females; mean APACHE II score, 30.2 [8.8]) developed shock (Table 2). Fifteen other patients who developed shock after the first 24 h of ICU stay were excluded. Protocol-based fluid management was used 455 times for 242 patients (39.5% of 612 patients), with 244 (53.6% of 455) positive responses. Protocol-based fluid management adherence, compared with nonadherence, was associated with sepsis diagnoses and hypertension as a comorbidity.

Table 2:
Patient characteristics

In the first 24 h of ICU stay, PBFM adherence, compared with nonadherence, was not associated with differences in baseline fluid balance. However, the former was associated with higher subsequent cumulative fluid balances, more vasoactive drug use, increased ventilator days, greater ICU LOS, and greater hospital LOS (Table 3). For septic patients, discordant antibiotic use and early antibiotic administration did not differ by PBFM use. Adjusted for age, sex, APACHE II score, comorbidity, and admission year, protocol use was associated with reduced ICU mortality (odds ratio [OR], 0.60; 95% confidence interval [95% CI], 0.39–0.94; P = 0.025) (Table 4) but not hospital mortality (OR, 0.82; 95% CI, 0.54–1.23; P = 0.369) (Table 5).

Table 3:
Clinical management and outcomes
Table 4:
Association of PBFM with ICU mortality
Table 5:
Association of PBFM with hospital mortality

For the detailed analysis of fluid challenges, out of the 293 fluid challenges done using the protocol for 113 patients in 2012, 216 (73.7%) fluid challenges had a cardiac output monitor present (Table 6). We analyzed the correlation between the change in PP and the change in SV after the first 250-mL fluid bolus and found a reasonable correlation between the two measures (Fig. 1; Pearson r = 0.48; P < 0.001), in line with prior published evidence (19). Using SV change as the reference parameter, sensitivity ranged from 58.7% to 79.1%, specificity ranged from 57.7% to 78.2%, agreement ranged from 64.4% to 75.0%, and κ showed fair agreement (23). Using protocol-defined volume response as the reference, sensitivity ranged from 51.5% to 89.2%, specificity ranged from 89.5% to 100.0%, agreement ranged from 66.2% to 93.5%, and κ ranged from 0.38 to 0.87 (i.e., fair to excellent agreement), with the best-performing parameter being PP change during two consecutive 250-mL boluses.

Table 6:
PP change and SV change characteristics of 216 fluid challenges done using PBFM in 2012
Fig. 1:
Scatter plot and line of best fit for paired measurements of percentage PP change and percentage SV change after a 250-mL fluid bolus (Pearson r = 0.48, P < 0.001).


Our results show that adherence to PBFM for shock within the first 24 h of ICU stay was associated with significantly improved ICU survival. The association, although in the same direction for hospital survival, did not achieve statistical significance. About half of the fluid challenges had positive responses. Analysis of secondary outcomes showed that protocol use was associated with increased ventilator days, ICU LOS, and hospital LOS.

Improved ICU survival with protocol use might be caused by more aggressive fluid loading (2, 13, 24). This was demonstrated by the greater cumulative fluid balances at 6, 12, and 24 h of ICU stay in the PBFM-adherent group compared with the nonadherent group. The median fluid balance at 24 h diverged by more than 1,000 mL, which was previously shown to be associated with large differences in mortality (2). However, a larger fluid volume by itself could not have been the only reason for improved survival because excessive fluids were previously shown to be deleterious (10, 11). A different explanation for improved survival might be that protocol use was associated with more appropriate fluid resuscitation. This was suggested by the observation that fluid challenges were approximately 50% positive and 50% negative for our patients, which is in line with prior research (6, 21). Protocol-based fluid management could have also worked by eliciting more thorough physician reviews, although we could not quantify these via the existing records.

An apparent trade-off between patient survival and resource utilization was present in our cohort. The higher fluid balances probably reflected necessary treatment for severely ill patients with shock (2). On the same note, more than 90% of the patients required vasoactive drugs within the first 24 h of ICU stay. Furthermore, protocol use was associated with increased ventilator days, ICU LOS, and hospital LOS. Given the increased fluid balances achieved by patients who underwent PBFM versus those who did not, we postulate that this impeded liberation from mechanical ventilation and perhaps subsequent convalescence among ICU survivors (9). In view of this trade-off, we believe that the statistically significant associations with decreased ICU mortality and the trend toward decreased hospital mortality remain important for further study.

The study has several strengths. First, we included patients with a broad range of diagnoses, extending the literature on fluid resuscitation to nonseptic cohorts. Second, we only considered very ill patients (mechanically ventilated patients who were also hypotensive or who required vasoactive drugs), thus minimizing the variability of ICU admission thresholds, and the latter’s effect on clinical outcomes. Third, we optimized the sensitivity of a 500-mL fluid challenge by separately infusing two consecutive 250-mL boluses. In particular, using protocol-defined volume response as the reference, the sensitivity and specificity of consecutive 250-mL boluses using the PP end point were excellent. Fourth, because our protocol required only arterial line blood pressure measurements and made cardiac output monitors optional, the study methods and results may be readily used for patients in many other ICUs, including resource-limited ones.

Nonetheless, our study has important limitations. First, it is an observational study that is vulnerable to residual confounding. For instance, protocol use and improved outcomes may be confounded by the presence of more conscientious clinical staff, although the large number of nurses (∼100) and physicians (∼200 intensivists and trainees) rotating through the ICU during the study period would have made such a factor unlikely. Second, patients with sepsis and hypertension (as a comorbidity) were more likely to receive PBFM, but these conditions were not associated with mortality in univariate or multivariate analyses. Third, our results may only apply to ICU patients who are severely compromised (mechanically ventilated and in shock) and may not be generalizable to other less-ill patients, including those not mechanically ventilated. Fourth, ED and ICU providers may have misclassified patients with and without sepsis, which would then mean overuse and underuse of antibiotics, respectively, but mortality in our patients was not associated with sepsis diagnosis. Fifth, we did not have data on any preexisting drug use (e.g., beta-blockers, statins), although this would likely not have affected the decision to use the protocol. Sixth, on pragmatic grounds, our protocol made SV monitors optional and, hence, we would not know the additional impact if the latter were mandatory.

Our findings showed that fluid management for shock can be facilitated by a pragmatic protocol, which we hope can spur more research in the optimal method of fluid resuscitation. We believe that the presence of a protocol may help avert haphazard management, and that recent evidence showing no benefit of protocolized shock management needs to be interpreted cautiously (1, 25). Concurrently, we demonstrated that focusing on fluid resuscitation for shock, without the need for complex bundles (14), could still be associated with improved outcomes. Important elements of our PBFM included physician-nurse cooperation and clear documentation of fluid challenge end points. These should be emphasized as part of standard care, which, if done well, has been shown to have outcomes comparable to treatment using a more invasive protocol for septic shock (1, 25). Additional studies on other patient populations (e.g., those who do not require mechanical ventilation, postoperative cases) will be required to prove wider applicability of the results. Furthermore, improved ICU survival may not always translate into overall improved hospital (and posthospital) survival. This may require development of post-ICU care programs and protocols to optimize the outcomes of ICU survivors.


In conclusion, we have shown that use of a pragmatic fluid management protocol was associated with significantly improved ICU survival among mechanically ventilated patients who developed shock within the first 24 h of ICU stay and who were admitted from the our ED to the ICU. Greater adherence to PBFM should be encouraged.


The authors thank Dr. Ma Thin Mar Win, Yong Loo Lin School of Medicine, National University of Singapore, for providing an independent statistical review.


1. Yealy DM, Kellum JA, Huang DT, Barnato AE, Weissfeld LA, Pike F, Terndrup T, Wang HE, Hou PC, LoVecchio F, et al.: A randomized trial of protocol-based care for early septic shock. N Engl J Med 370 (18): 1683–1693, 2014.
2. Lee SJ, Ramar K, Park JG, Gajic O, Li G, Kashyap R: Increased fluid administration in the first three hours of sepsis resuscitation is associated with reduced mortality: a retrospective cohort study. Chest 146 (4): 908–915, 2014.
3. Bozza FA, Carnevale R, Japiassu AM, Castro-Faria-Neto HC, Angus DC, Salluh JI: Early fluid resuscitation in sepsis: evidence and perspectives. Shock 34 (Suppl 1): 40–43, 2010.
4. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K, Angus DC, Brun-Buisson C, Beale R, et al.: Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 36 (1): 296–327, 2008.
5. Zhao H, Heard SO, Mullen MT, Crawford S, Goldberg RJ, Frendl G, Lilly CM: An evaluation of the diagnostic accuracy of the 1991 American College of Chest Physicians/Society of Critical Care Medicine and the 2001 Society of Critical Care Medicine/European Society of Intensive Care Medicine/American College of Chest Physicians/American Thoracic Society/Surgical Infection Society sepsis definition. Crit Care Med 40 (6): 1700–1706, 2012.
6. Michard F, Teboul JL: Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest 121 (6): 2000–2008, 2002.
7. Parker MM, Shelhamer JH, Bacharach SL, Green MV, Natanson C, Frederick TM, Damske BA, Parrillo JE: Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med 100 (4): 483–490, 1984.
8. Vincent JL: Let’s give some fluid and see what happens versus the mini-fluid challenge. Anesthesiology 115 (3): 455–456, 2011.
9. Durairaj L, Schmidt GA: Fluid therapy in resuscitated sepsis: less is more. Chest 133 (1): 252–263, 2008.
10. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, Moreno R, Carlet J, Le Gall JR, Payen D: Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 34 (2): 344–353, 2006.
11. Maitland K, Kiguli S, Opoka RO, Engoru C, Olupot-Olupot P, Akech SO, Nyeko R, Mtove G, Reyburn H, Lang T, et al.: Mortality after fluid bolus in African children with severe infection. N Engl J Med 364 (26): 2483–2495, 2011.
12. Paratz JD, Stockton K, Paratz ED, Blot S, Muller M, Lipman J, Boots RJ: Burn resuscitation—hourly urine output versus alternative endpoints: a systematic review. Shock 42 (4): 295–306, 2014.
13. Lin SM, Huang CD, Lin HC, Liu CY, Wang CH, Kuo HP: A modified goal-directed protocol improves clinical outcomes in intensive care unit patients with septic shock: a randomized controlled trial. Shock 26 (6): 551–557, 2006.
14. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M: Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 345 (19): 1368–1377, 2001.
15. Rady MY, Nightingale P, Little RA, Edwards JD: Shock index: a re-evaluation in acute circulatory failure. Resuscitation 23 (3): 227–234, 1992.
16. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 20 (6): 864–874, 1992.
17. De Backer D, Marx G, Tan A, Junker C, Van Nuffelen M, Huter L, Ching W, Michard F, Vincent JL: Arterial pressure-based cardiac output monitoring: a multicenter validation of the third-generation software in septic patients. Intensive Care Med 37 (2): 233–240, 2011.
18. Vincent JL, Weil MH: Fluid challenge revisited. Crit Care Med 34 (5): 1333–1337, 2006.
19. Monnet X, Letierce A, Hamzaoui O, Chemla D, Anguel N, Osman D, Richard C, Teboul JL: Arterial pressure allows monitoring the changes in cardiac output induced by volume expansion but not by norepinephrine. Crit Care Med 39 (6): 1394–1399, 2011.
20. Lau TS: On repeated screening tests. Biometrics 45 (3): 891–898, 1989.
21. Marik PE, Monnet X, Teboul JL: Hemodynamic parameters to guide fluid therapy. Ann Intensive Care 1 (1): 1, 2011.
22. Ostergaard M, Nilsson LB, Nilsson JC, Rasmussen JP, Berthelsen PG: Precision of bolus thermodilution cardiac output measurements in patients with atrial fibrillation. Acta Anaesthesiol Scand 49 (3): 366–372, 2005.
23. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33 (1): 159–174, 1977.
24. Jeon K, Shin TG, Sim MS, Suh GY, Lim SY, Song HG, Jo IJ: Improvements in compliance with resuscitation bundles and achievement of end points after an educational program on the management of severe sepsis and septic shock. Shock 37 (5): 463–467, 2012.
25. ARISE Investigators, ANZICS Clinical Trials Group, Peake SL, Delaney A, Bailey M, Bellomo R, Cameron PA, Cooper DJ, Higgins AM, Holdgate A, Howe BD, Webb SA, et al.: Goal-directed resuscitation for patients with early septic shock. N Engl J Med 371 (16): 1496–1506, 2014.

Intensive care units; ventilation; shock; hypotension; hemodynamics; fluid therapy; APACHE—Acute Physiology and Chronic Health Evaluation; CI—confidence interval; ED—emergency department; ICU—intensive care unit; IQR—interquartile range; LOS—length of stay; OR—odds ratio; PBFM—protocol-based fluid management; PP—pulse pressure; SD—standard deviation; SV—stroke volume

© 2015 by the Shock Society