Share this article on:

Urine Flow Is a Novel Hemodynamic Monitoring Tool for the Detection of Hypovolemia

Shamir, Micha Y. MD; Kaplan, Leonid MD; Marans, Rachel S. MD; Willner, Dafna MD; Klein, Yoram MD

doi: 10.1213/ANE.0b013e31820ad4ef
Technology, Computing, and Simulation: Research Reports
Chinese Language Editions

BACKGROUND: Noticeable changes in vital signs indicating hypovolemia occur only after 15% of the blood volume is lost. More sensitive variables (e.g., cardiac output, systolic pressure variation and its Δdown component) are invasive and difficult to obtain in the early phase of bleeding. Lately, a new technology for continuous optical measurements of minute-to-minute urine flow rates has become available. We performed a preliminary evaluation to determine whether urine flow can act as an early and sensitive warning of hypovolemia.

METHODS: Eleven patients (ASA physical status I–II) undergoing posterior spine fusion surgery were studied prospectively. Study variables included heart rate, blood pressure (systolic and diastolic), systolic pressure variation and Δdown, minute urinary flow, hemoglobin, blood and urinary sodium, and creatinine in the blood and urine. Urine flow rate was measured using URINFO 2000™ (FlowSense Medical, Misgav, Israel). After recording baseline variables, 10 mL/kg of the patient's blood was shed and a second set of variables was recorded. Subsequently, hypovolemia was reversed by infusing colloid solution (hetastarch 6%) followed by recording a third set of variables. These 3 observations were then compared.

RESULTS: An average of 614 ± 143 mL (mean ± SD) of blood was shed. During phlebotomy, the mean urine flow rate decreased from 5.7 ± 8 mL/min to 1.07 ± 2.5 mL/min. Systolic blood pressure and hemoglobin also decreased. Δdown increased. After rehydration, urine flow, blood pressure, and Δdown values returned to baseline. The hemoglobin concentration decreased whereas other variables did not change significantly.

CONCLUSION: Urine flow rate is a dynamic variable that seems to be a reliable indicator of changes in blood volume. These results justify further investigation.

Published ahead of print February 8, 2011 Supplemental Digital Content is available in the text.

From the Department of Anesthesiology and Critical Care Medicine, Hadassah– Hebrew University Medical Center, Jerusalem, Israel.

Authors' current affiliations are provided at the end of the article.

Supported by internal funding. Part of this study was done as part of the requirements of RSM for an MD degree from the Hadassah–Hebrew University Medical School, Jerusalem, Israel.

MYS and LK are co-first authors.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Micha Y. Shamir, MD, Department of Anesthesiology and Critical Care Medicine, Hadassah–Hebrew University Medical Center, POB 12000, Jerusalem 91120, Israel. Address e-mail to

Accepted November 24, 2010

Published ahead of print February 8, 2011

Hemorrhagic shock is among the leading causes of death after trauma.1 Early diagnosis of hypovolemia is thus of the utmost importance.2 Clinical diagnosis of hemorrhagic shock and measurements of its severity are based on traditional vital signs including heart rate, arterial blood pressure, respiratory rate, level of consciousness, and urine output (mL/h).3 Unfortunately, these variables change only after 15% of the estimated blood volume is lost and become strikingly apparent only after a 30% loss, which is a life-threatening situation.4 Hypovolemia can be diagnosed earlier using invasive monitoring tools such as pulmonary capillary wedge pressure, cardiac output, and arterial systolic pressure variation (SPV). Obtaining these variables and their interpretation is cumbersome and therefore are not considered standard of care at the early phase of patient evaluation.

Urine output is a sensitive variable reflecting the patient's effective blood volume and tissue perfusion. Urinary catheters are routinely inserted very early in the resuscitation phase in trauma victims so that urine output can be used as a diagnostic tool.5 Because small volumes are difficult to measure, initial information only becomes available 15 to 20 minutes after catheter insertion and the initial measurement of urine output is extrapolated to record the average value for urine output per hour (mL/h). This extrapolation can result in considerable over- or underestimation. A similar time period is needed to assess the immediate effect of medical intervention.

URINFO™ (Fig. 1; FlowSense Medical, Misgav, Israel) is a novel urine collecting and measurement system. It was originally introduced to assist nurses in measuring urine volume. This monitor incorporates dynamic optical flow-sensing technology, providing real-time continuous automated measurements of urine flow. Urine volume measurement errors are small because flow is optically measured. Also, because the measurement is continuous, minute-to-minute measurements of urine flow can be displayed. This is in contrast to traditional urinometers, which do not permit the display of minute urine flow.

Figure 1

Figure 1

The aim of this study was to evaluate whether minute urine flow rate can be used as a new variable that would provide early warning of hypovolemia.

Back to Top | Article Outline


This study was approved by the IRB of the Hadassah– Hebrew University Medical Center. All patients (or their legal guardians/relative surrogates) signed informed consent.

We studied 11 ASA physical status I and II patients scheduled for elective posterior spine fusion for idiopathic scoliosis using normovolemic hemodilution. Inclusion criteria included significant anticipated intraoperative hemorrhage (>4 vertebrae fusion) that would justify the use of hemodilution. Exclusion criteria included admission hemoglobin <11 g/dL; evidence of active infection; uncontrolled hypertension; significant cardiac, kidney, liver, or lung disease; arterial oxygen saturation <94%; and chronic use of diuretic or any drug that affects arterial blood pressure or heart rate.

Back to Top | Article Outline


Patients were premedicated with diazepam 0.15 mg/kg. Upon arrival to the operating room, the patients were connected to standard monitoring (electrocardiogram, noninvasive arterial blood pressure, pulse oximeter, and capnography). General anesthesia was induced with IV fentanyl 1 μg/kg, IV propofol 2 mg/kg, and IV vecuronium 0.1 mg/kg. After tracheal intubation, anesthesia was maintained with isoflurane 0.5% in an O2/N2O mixture at a 1:1 ratio. Patients' lungs were ventilated at a rate of 10 to 12 breaths/min and tidal volume of 6 to 8 mL/kg adjusted to keep end-tidal CO2 at approximately 40 mm Hg. After induction of anesthesia, the radial artery was cannulated using a 20-gauge catheter and a urinary catheter introduced to the urinary bladder. This catheter was connected to the URINFO 2000™ system for continuous measurements, which were downloaded to a laptop computer for analysis.

Back to Top | Article Outline


Study measurements included blood pressure (systolic and diastolic), heart rate, and urinary flow. Blood was obtained for hemoglobin, sodium, and creatinine measurements and urine samples were obtained for sodium and creatinine measurements (and fractional excretion of sodium was calculated). An invasive blood pressure waveform tracing was printed for off-line SPV and Δdown calculations. Δdown was calculated as the mean value of 3 positive pressure ventilator cycles. Urine flow was calculated as the mean value over 5 minutes after 5 minutes of no intervention allowing physiologic stabilization to be obtained. We needed a mean over 5 minutes because of our unique finding that urinary flow is not constant but exhibits significant minute-to-minute variability.

A complete set of measurements was recorded and laboratory samples were obtained in 3 stages: (1) baseline (after induction of general anesthesia); (2) after 10 mL/kg (approximately 15% estimated blood volume) of blood was shed as a free flow from the radial artery cannula (phlebotomy stage); and (3) after colloid infusion (hetastarch 6%) at a volume equal to the bloodshed volume (rehydration phase). Based on previous studies, patients were allowed 5 minutes before each set of measurements to reach physiologic stability.6,7 Shed blood was reinfused during the surgery at the anesthesiologist's discretion. If it was not needed during the surgery, it was reinfused at the conclusion of surgery. (Intraoperative donated blood is not accepted by the institution's blood bank.)

Back to Top | Article Outline

Statistical Analysis

Paired t tests were used to assess differences in variables between the periods: phlebotomy (after withdrawing 10 mL/kg blood) versus baseline, and rehydration (after colloid intravascular volume expansion) versus phlebotomy. The nonparametric Wilcoxon signed test was used where the distribution of variables was not normal. In this case, Monte Carlo significance values were calculated. A value of P ≤ 0.05 was considered statistically significant.

Back to Top | Article Outline


Eleven patients aged 20.9 ± 7.6 years (range, 13–37 years) participated in the study. The average weight was 58.9 ± 13.3 kg. Blood volume removed from the patients was 614 ± 143 mL.

Table 1 summarizes the changes in the different study variables at the 3 stages of the study. After blood removal, urine flow rate decreased from an average of 5.7 ± 8 mL/min before phlebotomy to an average of 1.07 ± 1.2 mL/min after (P = 0.002). The average Δdown increased significantly (P = 0.008) to a value >5 mm Hg, which is the cutoff value suggesting hypovolemia.8,9 Systolic blood pressure decreased 10.3 mm Hg (P = 0.029) and hemoglobin decreased 0.5 g/dL (P = 0.006); both were not clinically significant changes. After rehydration, urine flow rate increased to an average of 2.1 ± 2.5 mL/min. This change was prominent but not statistically significant, relative to the change in urine flow after phlebotomy. Arterial blood pressure increased, Δdown decreased, and hemoglobin decreased.

Table 1

Table 1

Figure 2 depicts a continuous urine flow chart of a single patient. This figure demonstrates the decrease in urine flow, which appears as early as 13 minutes after the phlebotomy started, approximately one-third of the total time needed for phlebotomy. It is also evident from this figure that urinary flow exhibits a minute-to-minute variation, which decreases in amplitude as blood is lost.

Figure 2

Figure 2

Back to Top | Article Outline


Decreased Urine Flow as a Sign of Hypovolemia

In this study, we evaluated the utility of using urine flow rate as an early sign of hypovolemia due to hemorrhage. Our results demonstrated significant changes in urine flow rate associated with bleeding. These findings are in agreement with those of Sondeen et al.10 who found that urine flow rate decreased significantly in conscious bleeding pigs when the amount of blood loss was between 10% and 20% of estimated blood volume. In addition, the first variable to change in relation to renal blood flow was urine flow. These changes were attributed to decreased renal perfusion during hypovolemia. We did not demonstrate a decrease in the fractional excretion of sodium (FENa) and it thus seems that the change in urine flow is not secondary to a change in renal function toward water preservation. These results are also in accordance with findings in pigs in which FENa remained unchanged until 40% of estimated blood volume was lost over 45 minutes.10 Similar results were found in conscious humans subjected to graded hemorrhage. In that study, urine flow rate decreased before arterial blood pressure and heart rate whereas FENa did not change at an estimated 10% to 20% blood volume loss.11 SPV and Δdown changed as early as urinary flow, thus flow is not a superior variable to monitor. However, SPV and Δdown require arterial cannulation, which is rarely used in the early stages of trauma resuscitation. Systolic blood pressure decreased significantly, but in clinical day-to-day work, this magnitude of change can be overlooked. Our results did not show a significant increase in urine flow after rehydration. This finding might suggest that urine flow can act as an early sign of bleeding or hypovolemia, but not as a tool to define the end of resuscitation. This study was not designed to verify this issue and further study is needed.

Back to Top | Article Outline

Human Hemorrhage Model

Studying hemorrhage in humans is extremely difficult. Alex Stone, in 1969, examined graded hemorrhage in conscious volunteers.11 His findings are unique and important, but it is impossible to use his protocol for research today because of current ethical considerations. Namely, there is no clinical situation in which numerous patients will bleed without the investigator's interference. We identified and used intraoperative blood donation and hemodilution as an experimental model illustrating human hemorrhage.9 While performing intraoperative hemodilution, a predetermined volume of blood is removed thus mimicking surgical or traumatic hemorrhage.12,13 There is no need to prove that hemorrhage occurred nor to estimate its magnitude because it is an inherent part of the procedure. The removed blood is later reinfused as needed and limited to 15% of estimated blood volume.

Two observations in this study support the validity of this model: SPV or its Δdown derivate, and the hemoglobin concentration. SPV and Δdown were proven to be superior indicators of hypovolemia.14,15 The mean value of Δdown after hemorrhage was 5.4 mm Hg in our patients, a value consistent with hypovolemia.8,9 The hemoglobin concentration changed as expected for acute blood loss, where hemoglobin concentration decreases significantly only after intravascular volume repletion by IV infusion or water redistribution from the extravascular compartment. The limitation of this model is that the maximal volume of lost blood is the upper limit of class I hemorrhage in which no significant hemodynamic changes are anticipated.16

Back to Top | Article Outline

Minute-to-Minute Urine Flow Variability

We noticed that minute-to-minute urine flow variability decreased during hemorrhage (Fig. 2). We believe this is the first description of this phenomenon. Variability disappears approximately half way through the bleeding process and returns after rehydration. Minute-to-minute variability in heart rate also decreased with hemorrhage and increased with sympathetic tone.17 In rabbits, decreased arterial blood pressure variability was found to reflect low circulating blood volume.18 In sheep, the phenomenon was also found to be related to the kidney because decreased heart rate variability during hemorrhage was associated with increased renal sympathetic nerve activity.19 In the present study, minute-to-minute urine flow variability was not equally prominent in all the patients. Further research is required to explore these findings. Urine flow variability may be a novel modality to assess hypovolemia.

Back to Top | Article Outline

Study Limitations

Systolic blood pressure and hemoglobin decreased slightly after 15% blood loss, an unanticipated change in grade I hemorrhage. Because systolic blood pressure changes were not clinically significant (<10%), we do not consider them a meaningful clinical change. The same is true for the 0.5 g/dL decrease in hemoglobin after phlebotomy. Alternatively, the combination of preoperative fasting and removing 15% of the blood volume (upper volume limit for class I hemorrhagic shock) might result in a deeper hypovolemia than intended. In combination with general anesthesia, some susceptible patients might react in a more pronounced way.

This study was designed as a feasibility study and thus the sample size is small. Because each patient in our study acted as his/her own control, we find the sample size sufficient. Nevertheless, a larger-scale study is needed and should be designed to measure the usefulness of urine flow rate to monitor the effectiveness of intravascular volume-replacement treatment.

Our data show that urine flow rate can be used as a sensitive sign of hemorrhage. Further trials are needed to study the topic in animals (whereby we can subject the animals to significant hemorrhage) and in larger-scale human studies.

Back to Top | Article Outline


Micha Y. Shamir, MD, is currently affiliated with the Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami Miller School of Medicine, Miami, FL; Leonid Kaplan, MD, is currently affiliated with the Department of Orthopedic Surgery, Hadassah–Hebrew University Medical Center; Rachel S. Marans, MD, is currently affiliated with the Department of Pediatrics, Hadassah– Hebrew University Medical Center; Dafna Willner, MD, is currently affiliated with the Department of Anesthesiology and Critical Care Medicine, Hadassah–Hebrew University Medical Center, Jerusalem, Israel; and Yoram Klein, MD, is currently affiliated with the Department of Surgery, Trauma Unit, Kaplan Medical Center, Rehovot, Israel.

Back to Top | Article Outline


Name: Micha Y. Shamir, MD

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Micha Y. Shamir, MD, has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Leonid Kaplan, MD

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Leonid Kaplan, MD, has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Rachel S. Marana, MD

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Rachel S. Marans, MD, has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Dafna Willner, MD

Contribution: This author helped design the study, conduct the study, and write the manuscript.

Attestation: Dafna Willner, MD, has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Yoram Klein, MD

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Yoram Klein, MD, has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Back to Top | Article Outline


1. Cocchi MN, Kimlin E, Walsh M, Donnino MW. Identification and resuscitation of the trauma patient in shock. Emerg Med Clin North Am 2007;25:623–42
2. Gutierrez G, Reines HD, Wulf-Gutierrez ME. Clinical review: hemorrhagic shock. Crit Care 2004;8:373–81
3. Dutton RP. Shock and trauma anesthesia. Anesthesiol Clin North America 1999;17:83–97
4. Spahn DR, Cerny V, Coats TJ, Duranteau J, Fernández-Mondéjar E, Gordini G, Stahel PF, Hunt BJ, Komadina R, Neugebauer E, Ozier Y, Riddez L, Schultz A, Vincent JL, Rossaint R. Management of bleeding following major trauma: a European guideline. Crit Care 2007;11:R17
5. American College of Surgeons Committee on Trauma. Shock. In: American College of Surgeons. Advanced Trauma Life Support for Doctors. 6th ed. Chicago: American College of Surgeons, 1997:101–7
6. Rooke GA, Schwid HA, Shapira Y. The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation. Anesth Analg 1995;80:925–32
7. Tavernier B, Makhotine O, Lebuffe G, Dupont J, Scherpereel P. Systolic pressure variation as a guide to fluid therapy in patients with sepsis-induced hypotension. Anesthesiology 1998;89:1313–21
8. Perel A, Pizov R, Cotev S. Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded hemorrhage. Anesthesiology 1987;67:498–502
9. Shamir M, Eidelman LA, Floman Y, Kaplan L, Pizov R. Pulse oximetry plethysmographic waveform during changes in blood volume. Br J Anaesth 1999;82:178–81
10. Sondeen JL, Gonzaludo GA, Loveday JA, Deshon GE, Clifford CB, Hunt MM, Rodkey WG, Wade CE. Renal responses to graded hemorrhage in conscious pig. Am J Physiol 1990;259:R119–25
11. Stone AM, Stahl WM. Renal effects of hemorrhage in normal man. Ann Surg 1970;172:825–36
12. Shander A, Rijhwani TS. Acute normovolemic hemodilution. Transfusion 2004;44:26s–34s
13. Monk TG. Acute normovolemic hemodilution. Anesthesiol Clin North America 2005;23:271–8
14. Ornstein E, Eidelman LA, Drenger B, Elami A, Pizov R. Systolic pressure variation predicts the response to acute blood loss. J Clin Anesth 1998;10:137–40
15. Pizov R, Ya'ari Y, Perel A. Systolic pressure variation is greater during hemorrhage than during sodium nitroprusside-induced hypotension in ventilated dogs. Anesth Analg 1988;67:170–4
16. Convertino VA, Cooke WH, Holcomb JB. Arterial pulse pressure and its association with reduced stroke volume during progressive central hypovolemia. J Trauma 2006;61:629–34
17. Proctor KG, Atapattu SA, Duncan RC. Heart rate variability index in trauma patients. J Trauma 2007;63:33–43
18. Egi A, Kawamoto M, Kurita S, Yuge O. Systolic arterial pressure variability reflects circulating blood volume alterations in hemorrhagic shock in rabbits. Shock 2007;28:733–40
19. Batchinsky AI, Cooke WH, Kuusela TA, Jordan BS, Wang JJ, Cancio LC. Sympathetic nerve activity and heart rate variability during severe hemorrhagic shock in sheep. Auton Neurosci 2007;136:43–51
© 2011 International Anesthesia Research Society