CJA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Résumé de cet Article
Right arrow Full Text (PDF)
Right arrow Submit a scholarly reply
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ezri, T.
Right arrow Articles by Shimonov, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ezri, T.
Right arrow Articles by Shimonov, M.
Canadian Journal of Anesthesia 53:153-158 (2006)
© Canadian Anesthesiologists' Society, 2006
Regional Anesthesia and Pain

Phosphate salt bowel preparation regimens alter perioperative acid-base and electrolyte balance

[Les préparations intestinales aux sels phosphatés modifient l’équilibre périopératoire acide-base et électrolytique]

Tiberiu Ezri, MD*,{dagger}, Emma Lerner, MD{ddagger}, Michael Muggia-Sullam, MD§, Benjamin Medalion, MD, Alexander Tzivian, MD, Abraham Cherniak, MD**, Peter Szmuk, MD{dagger}{dagger} and Mordechai Shimonov, MD{ddagger}{ddagger}

* From the Departments of Anesthesia,
{ddagger} Internal Medicine "B",
§ Surgery "B",
Cardiothoracic Surgery,
** Urology, Surgery "A",
{dagger}{dagger} Anesthesiology, The University of Texas Medical School at Houston, TX;
{ddagger}{ddagger} Surgery "A", Wolfson Medical Center, Holon, Affiliated to Sackler School of Medicine, Tel Aviv University, Israel; and the
{dagger} Outcomes ResearchTM Institute, University of Louisville, Kentucky, USA.

Address correspondence to: Dr. Peter Szmuk, Department of Anesthesiology, The University of Texas Medical School at Houston, 6431 Fannin, MSB 5.020, Houston, Texas 77030, USA. E-mail: peter.szmuk{at}uth.tmc.edu


   Abstract
TOP
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
Background: Hyperphosphatemic acidosis and severe electrolyte disturbances caused by phosphate salts (PO) used for mechanical bowel preparation have been described in occasional case reports prior to bowel resection surgery. We hypothesized that PO used preoperatively for bowel preparation may cause more pronounced acid base and electrolyte changes than polyethylene glycol (PG).

Methods: Forty American Society of Anesthesiologists physical status II–III patients were randomly allocated to receive either PO or PG for bowel preparation before intra-abdominal surgery (bowel resection or other major elective intra-abdominal surgeries). Measurements of pH, base deficit, blood gases, lactate, hemoglobin, calcium, magnesium, potassium and phosphorus were undertaken before the laxative administration, intraoperatively, and postoperatively.

Results: Preoperative demographic, hemodynamic and laboratory data were similar in the two groups. Intraoperative calcium (8.4 [0.6] vs 9 [0.5] mg·dL–1) and pH (7.35 [0.04] vs 7.41 [0.03]) were lower, while lactate (1.3 [0.4] vs 0.9 [0.3] mmol·L–1) was higher with PO. Postoperative calcium, magnesium and potassium were lower (8 [0.5] vs 8.9 [0.2] mg·dL–1, 1.68 [0.3] vs 1.8 [0.4] and 3.5 [0.36] vs 3.7 [0.33] mEq·L–1 respectively) while phosphorus (4.1 [0.3] vs 3.3 [0.2] mEq·L–1) was higher with PO. A higher percentage of abnormal values for calcium, potassium, phosphorus and base deficit (66% vs 33%, 25% vs 10%, 19% vs 2% and 28.3% vs 5% respectively) were observed with PO.

Conclusions: Calcium and magnesium changes were more pronounced in patients who received PO for bowel preparation.


   Introduction
TOP
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
METABOLIC acidosis may occur following major abdominal surgery (bowel resection, pancreatectomy, etc.) and may be caused by multiple mechanisms: "dilution" acidosis due to infusion of large volumes of saline solution,1,2 lactic acidosis due to hypovolemia and/or low perfusion state (i.e., prolonged organ ischemia, decreased lactate metabolism by liver) and ketoacidosis due to either prolonged fasting,3 alcohol intoxication4 or diabetes. A rare reason for perioperative acidosis is hyperphosphatemia.

Hyperphosphatemic acidosis caused by phosphate salts (PO) used for mechanical bowel preparation has been described in occasional case reports prior to bowel resection surgery.5 However, this problem is often overlooked by many physicians.6

Even in volumes as low as 90 mL, PO administration for bowel preparation may be associated with severe metabolic disorders (i.e., severe hyperphosphatemic acidosis), coma, hypovolemia, serious electrolyte disturbances including hyperphosphatemia, hypokalemia, hypomagnesemia, hypocalcemia (tetany has been described), hypernatremia, acute renal failure or even death.5,7–11 These disturbances are aggravated by preexisting congestive heart failure, renal function impairment, liver failure and slow bowel elimination therefore, in these patients, the use of polyethylene glycol (PG) instead of PO is indicated.11,12

We hypothesized that PO used preoperatively for bowel preparation may cause more pronounced acid base and electrolyte changes than PG.


   Methods
TOP
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
The study was conducted at the Wolfson Medical Center. The Institutional Review Board Committee approval and each patient’s informed consent were obtained before starting the study.

Primary outcomes of the study were perioperative changes in the two groups in base excess/deficit, pH, lactate, phosphate, calcium, magnesium and potassium serum levels during all time point measurements.

The sample size for this study was calculated as follows: although the apparent magnitude of change in acid base and electrolyte balance that might be considered to be problematic and clinically relevant would be any value which is higher or lower that the normal range of values, we defined a difference between groups of 5 mEq·L–1 for sodium, 0.05 units for pH, 5 mEq·L–1 for base excess/deficit, 0.5 mmol·L–1 for lactate, 0.3 mEq·L–1 for magnesium, 0.5 mEq·L–1 for potassium, 0.5 mg·dL–1 for serum inorganic phosphorus and 1 mg·dL–1 for total serum calcium, to be clinically important. This is because we considered that the trend of change itself and the amount of change and not only the absolute value is also important. To achieve an 80% power to detect such a difference with an {alpha} of 0.05, 17 patients were required in each group. To account for possible incomplete data acquisition, we therefore enrolled 20 patients per group (age ≥ 40 yr); subjects undergoing elective open bowel resection or other major elective intra-abdominal surgeries, which per routine necessitate bowel preparation.

Excluded from the study were patients with preexisting metabolic acidosis that was diagnosed before the bowel preparation (i.e., renal failure induced metabolic acidosis), decompensated congestive heart failure, impaired renal function (i.e., serum creatinine > 1.5 mg·dL–1), liver failure or other reasons for hyperphosphatemia (i.e., rabdomyolysis, tumour lysis, etc.). Under these circumstances, phosphate administration is contraindicated.

Patients were fasted from the night before surgery; however, on the morning of surgery, they were allowed to take their usual medications (i.e., antihypertensive drugs) with a sip of water.

In a prospective, single blind trial, 40 patients were randomly allocated with the closed envelope method to two study groups of 20 patients each, according to the type of regimen used for bowel preparation: group PG received bowel preparation with oral PG solution (2,700 mL total volume – one glass every ten minutes) and those in group PO received oral sodium phosphate solution two flacons (90 mL) as cathartic, and additional water as prescribed. The bowel preparation treatment was randomly ordered and carried out the night before surgery by the surgeon on call, who was not involved with the study. The anesthesiologist in charge of the blood sampling was "blinded" to the type of bowel treatment the patient received. Patients continued to have oral fluids until 12 p.m. the night before surgery. On the morning of surgery, iv lactated Ringer’s solution was started at a rate of 100 mL·hr–1 as maintenance fluid therapy.

In diabetic patients the preoperative blood glucose was kept between 100 to 200 mg·dL–1.

Anesthesia induction and maintenance were standardized: midazolam 1 mg iv, fentanyl 1 µg·kg–1 iv prior to induction (and then, aliquots of 50 µg iv as required), thiopental – "sleep" dose, rocuronium (provided bowel obstruction was ruled-out, in which case succinylcholine was employed for intubation in a rapid sequence manner), isoflurane (0.5 – 1.5 MAC) and 60% N2O in oxygen. Intraoperatively, all patients were ventilated in a volume mode (tidal volume 7 mL·kg–1 and respiratory rate 10 breaths·min–1).

In addition to routine monitoring, direct arterial pressure, central venous pressure, urine output and nasopharyngeal temperature were measured. Hypothermia was avoided with fluid warmers and forced – air warmers from the moment the patient was brought into the operating room. Patient’s tracheas were extubated according to standard criteria, when hemodynamically stable, ongoing surgical bleeding was excluded, core temperature was > 35°C and arterial pH > 7.20. Fluid input was standardized in both groups. Patients received lactated Ringer’s solution at a rate of 10 mL·kg–1·hr–1 (including fasting, intraoperative maintenance and third spacing). Blood was transfused according to the American Society of Anesthesiologists guidelines13 which state that blood transfusion is rarely indicated when the hemoglobin (Hb) concentration is > 10 g·dL–1 and is almost always indicated when it is < 6 g·dL–1, especially when the anemia is acute. The determination of whether Hb concentrations of 6 to 10 g·dL–1 justify blood transfusion was based on the patient’s risk for complications of inadequate oxygenation (i.e., we tried to keep Hb at 10 g·dL–1 or higher in patients suffering from ischemic heart disease).

Postoperative fluid replacement consisted of lactated Ringer’s solution 100 mL·hr–1, and if necessary, blood and fluid boluses were administered according to the above criteria. Diabetic patients with a blood glucose > 200 mg·dL–1 received a 5% glucose solution with a continuous infusion of insulin given at a rate of 2 IU·100 mL–1·hr–1. Potassium was added with a serum level < 3.5 mEq·L–1. Postoperatively, patients were managed in the postanesthesia care unit for at least four hours.

Measurements
Arterial blood samples were drawn in a preheparinized syringe. Standard bicarbonate, base deficit or excess, pH, PaCO2, PaO2, Hb, serum lactate, inorganic phosphorus, calcium (total), magnesium, chloride, sodium, potassium, urea, and glucose were analyzed before starting the bowel preparation (baseline), every hour during surgery and one, eight and 24 hr postoperatively. Measurements were performed with the Olympus Au 2700 analyzer system for clinical chemistry (Olympus Diagnostica GmbH, Hamburg, Germany). Mean arterial pressure, heart rate, central venous pressure, urine output and temperature were recorded at the same time points.

Data analysis
We performed a univariate analysis, where continuous variables with distribution significantly differing from normal were compared using the Mann-Whitney U or median tests. Categorical data were described using frequency counts and percentages and compared using Chi-square (with Yates correction) or Fisher’s exact tests, as appropriate. We performed two multi-variable analyses. Both were repeated measure general linear models. In the first multivariable analysis, the effect of study group, total fluid administered (which may affect by dilution the electrolytes measurements), age and sex were studied as predictors for changes in phosphorus, magnesium, calcium and potassium. In the second multivariable analysis, study group, total volume of lactated Ringer’s solution, normal saline infused and diabetes (all three, potentially affecting the acid-base balance) were analyzed as predictors for changes in pH, base excess/deficit and lactate. Secondary outcomes (during all time point measurements) were a need for mechanical ventilation due to a pH < 7.20 and any complication related to acidosis or electrolyte disturbances (i.e., arrhythmias). The effect of study group, age, sex, and amount of fluid administered, on the outcome variables were also studied.

Data analysis was performed using the SPSS for Windows (SPSS Inc, Chicago, IL, USA). All tests were considered significant at P < 0.05.


   Results
TOP
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
Preoperative data
Patients were scheduled for open intra-abdominal surgeries including bowel resection for tumours and radical prostatectomy. Demographic data presented in Table IGo were similar between groups. The incidence of ischemic heart disease, hypertension and hyper-lipidemia was similar, as was the American Society of Anesthesiologists physical status. Preoperative hemo-dynamic and laboratory data were also similar (Tables IGo and IIGo).


View this table:
[in this window]
[in a new window]
  		TABLE I Demographic data
 

View this table:
[in this window]
[in a new window]
  		TABLE II Intraoperative hemodynamics, urine output and temperature
 
Intraoperative data
Duration of surgery, the amount of intraoperative fluid and blood administered, and urinary output were similar (Table IGo). Intraoperative hemodynamic data were also similar at corresponding times (Table IIGo). The laboratory data with significant differences between the groups are presented in Table IIIGo.


View this table:
[in this window]
[in a new window]
  		TABLE III Univariate and multivariable analysis of laboratory data*
 
Intraoperative calcium and pH were lower, while base deficit and lactate were higher in the PO group (Table IIIGo). At the end of surgery the sodium, chloride, urea, glucose, arterial blood PaCO2 and arterial blood PaO2 were normal and similar between the groups (data not presented). Hemoglobin was 11.8 ± 1 vs 11.3 ± 1 mg·dL.

Postoperative data
Postoperative potassium, calcium and magnesium were lower, while phosphorus was higher in the PO group (Table IIIGo). The laboratory data were normal and similar in the two groups 24 hr after surgery, except for the serum inorganic phosphorus that was significantly higher in the PO group (Table IIIGo and FigureGo). Only three patients (in the PO group) had a pH < 7.30, two hours postoperatively. This was managed with boluses of fluids. No patient required postoperative mechanical ventilation as a result of acidosis.


Figure 1
View larger version (30K):
[in this window]
[in a new window]
  		FIGURE Perioperative serum inorganic phosphorus measurements. Group 1 = patients treated with sodium phosphate; Group 2 = patients treated with polyethylene glycol.

 
A higher percentage (of the total number of measurements) of values outside the normal range for calcium, potassium, phosphorus and base deficit was observed with the PO compared to the PG group (66% vs 33%, 25% vs 10%, 19% vs 2% and 28.3% vs 5% respectively).

Multivariable regression analysis showed temporal changes of pH, base deficit and lactate (P = 0.005), and calcium and magnesium (P = 0.01) within the groups and changes between the groups for calcium and magnesium (P = 0.05), (Table IIIGo). Multivariable analysis with repeated measure general linear models revealed a difference in the temporal changes of the base deficit, pH, lactate, calcium and magnesium between groups (Table IIIGo).

Although most of the averages were within the normal range, temporal changes in pH, base deficit, lactate, calcium and magnesium occurred within the two groups. Calcium and magnesium changed significantly between the two groups. No patient had a pH < 7.20, which would have mandated postoperative mechanical ventilation according to the protocol. No complications related to acidosis or electrolyte changes were observed.


   Discussion
TOP
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
Bowel preparation is routine in many abdominal surgeries. The preoperative preparation generally includes mechanical cleansing and antibiotic administration. The approach for mechanical cleansing had changed during recent years, from a three-day regimen consisting of low residual diet combined with laxatives and enemas, to a one-day regimen of isotonic solutions containing balanced salt solution such as PG, or sodium phosphate preparations like PO.14

A major impediment to efficient bowel preparation is noncompliance of patients. Although well tolerated by the patients, sodium PO solutions carry the risk of electrolyte disturbances that can aggravate preexisting renal or liver failure or reciprocally, the above mentioned preexisting diseases might aggravate the toxicity caused by the PO.11,12 We did not assess patient compliance with either preparation.

Our study demonstrated significant temporal changes in pH, base deficit, lactate, calcium and magnesium within the two groups. Calcium and magnesium changed significantly between the two groups. Although both preparations produce their cathartic activity by osmotic effect in the gut with a shift of fluid and electrolytes across the gut wall into the gut lumen, the probability of electrolyte disturbances differs.

Tan et al.8 reported six cases of severe electrolyte and metabolic derangements due to sodium phosphate bowel preparation: two patients had delayed awakening from general anesthesia, and four patients suffered life-threatening consequences. We did not encounter any delay in recovery from anesthesia, nor was there any life-threatening complication. This may be explained by the short-term bowel preparation of our patients, as well as the rigorous intraoperative hemodynamic monitoring and fluid replacement.

We found that the use of PO may alter the normal acid base balance of the patients more than with PG. In the PO group, pH values and base excess were significantly lower compared to the PG group (Table IIIGo). This difference cannot be explained by any single intraoperative event, considering that the fluid input and urine output where not different. Hill et al.15 demonstrated that patients treated with oral sodium phosphate have depleted intracellular potassium. These patients are at risk of hypokalemia after this treatment. Others have described that the occurrence of hyper-phosphatemia, hypocalcemia, hypomagnesemia and hypokalemia was common with the use of PO rather than with PG.8–10 In our series, the preoperative electrolyte status of the patients in both groups was similar. However, the persistent decrease in postoperative magnesium is related to the use of PO. This was also the case with the calcium, which resulted in a similar decreasing pattern during the first eight postoperative hours. Postoperative potassium levels were also lower in the PO group. In contrast to what was anticipated, hyperphosphatemia was not observed, presumably due to exclusion from the study of patients with heart, kidney and liver failure in whom hyperphosphatemia has been repeatedly reported following administration of PO.11,12

The changes in electrolytes were not long lasting, as they returned to normal 24 hr after surgery. Although we encountered significantly more pronounced peri-operative acid-base balance and electrolyte changes in patients who received sodium phosphate for bowel preparation before abdominal surgery, there were no complications related to those changes in either of the groups.

In a study of carefully selected elderly patient groups comparing PO with PG before colonoscopy, Thomson et al.16 concluded that sodium phosphate was a safe and effective bowel preparation, in spite of transient, mild changes in serum phosphorus and potassium.

One limitation of our study is that we cannot extrapolate our results to a high-risk patient group. However, it would be unethical to undertake such a study in at risk populations (patients suffering from heart, renal and hepatic failure), considering that the use of sodium phosphate solutions is contraindicated in those patients.

We conclude that in patients requiring bowel preparation prior to abdominal surgery, temporal changes in pH, base deficit, lactate, calcium and magnesium occurred within the sodium phosphate and polyethylene glycol groups. Calcium and magnesium changes were more pronounced in the sodium phosphate group.


   Footnotes
 
Accepted for publication July 13, 2005. Revision accepted September 9, 2005.


   References
TOP
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Waters JH, Miller LR, Clack S, Kim JV. Cause of metabolic acidosis in prolonged surgery. Crit Care Med 1999; 27: 2142–6.[Medline]

2 Waters JH, Gottlieb A, Schoenwald P, Popovich MJ, Sprung J, Nelson DR. Normal saline versus lactated Ringer’s solution for intraoperative fluid management in patients undergoing abdominal aortic aneurysm repair: an outcome study. Anesth Analg 2001; 93: 817–22.[Abstract/Free Full Text]

3 Valencia I, Pfeifer H, Thiele EA. General anesthesia and the ketogenic diet: clinical experience in nine patients. Epilepsia 2002; 43: 525–9.[Medline]

4 Hitender J, Sushil B, Shailender S. Alcohol induced ketoacidosis, severe hypoglycemia and irreversible encephalopathy. Med Sci Monit 2002; 8: 77–9.

5 Pitcher DE, Ford RS, Nelson MT, Dickinson WE. Fatal hypocalcemic, hyperphosphatemic, metabolic acidosis following sequential sodium phosphate-based enema administration. Gastrointest Endosc 1997; 46: 266–8.[Medline]

6 Hookey LC, Depew WT, Vanner S. The safety profile of oral sodium phosphate for colonic cleansing before colonoscopy in adults. Gastrointest Endosc 2002; 56: 895–902.[Medline]

7 Markowitz GS, Nasr SH, Klein P, et al. Renal failure due to acute nephrocalcinosis following oral sodium phosphate bowel cleansing. Hum Pathol 2004; 35: 675–84.[Medline]

8 Tan HL, Liew QY, Loo S, Hawkins R. Severe hyper-phosphataemia and associated electrolyte and metabolic derangement following the administration of sodium phosphate for bowel preparation. Anaesthesia 2002; 57: 478–83.[Medline]

9 Shaoul R, Wolff R, Seligmann H, Tal Y, Jaffe M. Symptoms of hyperphosphatemia, hypocalcemia, and hypomagnesemia in an adolescent after the oral administration of sodium phosphate in preparation for a colonoscopy. Gastrointest Endosc 2001; 53: 650–2.[Medline]

10 Lieberman DA, Ghormley J, Flora K. Effect of oral sodium phosphate colon preparation on serum electrolytes in patients with normal serum creatinine. Gastrointest Endosc 1996; 43: 467–9.[Medline]

11 Nir-Paz R, Cohen R, Haviv YS. Acute hyperphosphatemia caused by sodium phosphate enema in a patient with liver dysfunction and chronic renal failure. Ren Fail 1999; 21: 541–4.[Medline]

12 Ferguson KH, Neil JJ, Morey AF. Mechanical and anti-biotic bowel preparation for urinary diversion surgery. J Urol 2002; 167: 2352–6.[Medline]

13 Practice guidelines for blood component therapy: a report by the American Society of Anesthesiologists Task Force on blood component therapy. Anesthesiology 1996; 84: 732–47.[Medline]

14 Rubesin SE, Levine MS, Laufer I, Herlinger H. Double-contrast barium enema examination technique. Radiology 2000; 215: 642–50.[Abstract/Free Full Text]

15 Hill AG, Teo W, Still A, Parry BR, Plank LD, Hill GL. Cellular potassium depletion predisposes to hypokalaemia after oral sodium phosphate. Austr N Z J Surg 1998; 68: 856–8.

16 Thomson A, Naidoo P, Crotty B. Bowel preparation for colonoscopy: a randomized prospective trail comparing sodium phosphate and polyethylene glycol in a pre-dominantly elderly population. J Gastroenterol Hepatol 1996; 11: 103–7.[Medline]




This article has been cited by other articles:


Home page
 Canadian J. AnesthesiaHome page
M. Caswell
Phosphate salt bowel preparation regimens alter perioperative acid-base and electrolyte balance.
Can J Anesth, September 1, 2006; 53(9): 961 - 961.
[Full Text] [PDF]

Home page
 Canadian J. AnesthesiaHome page
T. Ezri, P. Szmuk, and M. Muggia-Sullam
REPLY
Can J Anesth, September 1, 2006; 53(9): 961 - 961.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Résumé de cet Article
Right arrow Full Text (PDF)
Right arrow Submit a scholarly reply
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ezri, T.
Right arrow Articles by Shimonov, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ezri, T.
Right arrow Articles by Shimonov, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS