Abstract
Background
Magnesium (Mg++) deficiency can result in life-threatening complications. The incidence of hypomagnesemia, as well as any coexisting hypokalemia and Electrocardiography (ECG) abnormalities, was studied in patients undergoing major gastrointestinal (GI) surgeries.
Methods
This observational study on 51 consecutive adult Intensive Care Unit (ICU) patients recorded serum Mg++ and serum potassium (K+) levels, and 12 lead ECGs, preoperatively and postoperatively, at 48 h and 72 h. Paired “t” test, Pearson Correlation Coefficient and chi-square test were used to statistically assess the difference, correlation, and association between serum Mg++, serum K+, and abnormal ECGs, respectively.
Results
Mean values for serum Mg++ were 1.72 mg/dl and 1.71 mg/dl on day 2 and 3 postops, respectively, while for serum K+ it was 4.14 meq/l and 4.02 meq/l. The incidence of postop hypomagnesemia was 52.9% with a 95% confidence interval (39.2–66.2) on Day 2 and 47.1%, with a 95% confidence interval (33.7–60.7) on Day 3. The incidence of coexisting hypokalemia was 33.3% on Day 2 and 29.2% on Day 3. There was no significant difference between pre and postop serum Mg++ and serum K+ values. The incidence of abnormal ECG was 33.3% on Day 2 postop and 28.6% on Day 3 and had a significant association with incidence of hypomagnesemia on Day 2 (P = 0.02).
Conclusion
Incidence of hypomagnesemia showed no significant difference pre and postoperatively. A significant association was present between the incidence of hypomagnesemia with abnormal ECG on the second postop day, but this was not found significant when compared with cases of hypomagnesemia with coexisting hypokalemia.
Keywords: GI surgery, Hypomagnesemia, Hypokalemia, Abnormal ECG, Postoperative ICU
Introduction
Magnesium (Mg++) is one of the major intracellular cations. It is vital for sustaining neuromuscular activity and acts as a cofactor for various enzymes.1 It is the second most abundant intracellular cation and the fourth most common cation in the body.2 Its deficiency is far more pronounced than is often noted in a susceptible subset of patients. Low levels of serum Mg++ can result in Electrocardiography (ECG) changes that can later develop life-threatening cardiac dysrhythmias with critically low levels.3 Hypomagnesemia can also often coexist with hypokalemia.
Mg++ homeostasis involves the kidney, small bowel, and bone. Most of the absorption of this electrolyte takes place in the jejunum and ileum,4 via both a passive paracellular mechanism and an active transport process. The paracellular mechanism accounts for 90% of intestinal Mg++ absorption. Any gastrointestinal disturbance in the form of ileus or major gut surgeries can affect this process.5
Hypokalemia is a common event in hypomagnesemia patients, occurring in 40–60% of cases.6 Magnesium deficiency is frequently associated with hypokalemia due to increased distal potassium secretion via the release of the magnesium-mediated inhibition of renal outer medullary potassium (ROMK) channels.7
Serum total Mg levels below 0.7 mmol/l is associated with electrocardiographic changes like ST-segment depression, flattened T waves, and prolongation of PR and QT/QTc intervals.8 Life-threatening arrhythmias associated with hypomagnesemia can occur at much lower levels and could include atrial fibrillation, premature ventricular contractions, ventricular tachycardia, and ventricular fibrillation.9
This study attempted to uncover the incidence of hypomagnesemia in postoperative gastrointestinal (GI) surgery patients, which hitherto goes undiscovered commonly due to the unavailability of means for analyzing this electrolyte in hospital setups. This at times induces empirical treatment of assumed hypomagnesemia based on clinical signs and symptoms like generalized muscle weakness, cramps, apathy, delirium, refractory hypokalemia, or major cardiac manifestations like supraventricular/ventricular arrhythmias. The secondary objectives of this study were also to detect the presence of coexisting hypokalemia, as well as to note any abnormal ECG patterns in these patients.
Although there are studies that show depletion of magnesium levels in the postoperative period after major surgery,10 there are no studies available linking hypomagnesemia with hypokalemia and abnormal ECG in postop GI surgical patients. Hence this study could go a long way in sensitizing the treating team to the importance of serum Mg++ levels, which, if found significantly low, can result in increased mortality and morbidity.
Materials and methods
This was projected to be an observational study of 100 consecutive adult patients in the Intensive Care Unit (ICU) of a tertiary care teaching hospital who underwent a major gut surgery and had a minimum postoperative ICU stay of 72 h. The sample size was based on previous studies showing the incidence of hypomagnesemia in ICU as around 60%. The confidence levels of 95% showed sample size as 97 patients, which was rounded off to 100 patients. In this study, fifty-one patients meeting the inclusion criteria could be studied in the given time period of 24 months from Oct 2016 to Sep 2018. The study was approved by the Institutional Ethics Committee. A written informed consent was taken from all patients.
A blood sample for serum magnesium was drawn on the morning of surgery for a baseline preoperative value and then again in the postoperative period at Day 2 and Day 3 after surgery for all the patients. Blood samples for serum potassium levels were also taken at the preop stage and at Days 2 and 3, respectively, during the postop period. Postop day 1 was not taken into account on the assumption that it would take the gut 24–48 h to recover postop. Similarly, a 12 lead ECG was also recorded in the postoperative period up to Day 3, apart from continuous ongoing ECG monitoring during the ICU stay. This was compared to a preop baseline 12 lead ECG taken on the morning of surgery, and any significant changes were noted.
Normal serum magnesium value (reference range) is 0.7–1.0 mmol/L or 1.7–2.4 mg/dl.6 Any value below 1.7 mg/dl was considered as hypomagnesemia. Normal serum potassium reference range was taken as 3.5–5.5 meq/L or mmol/l.
The following ECG changes were considered as abnormal or reflecting cardiac dysrhythmias in cases with coexisting hypomagnesemia and hypokalemia: unexplained tachycardia (Heart Rate > 120/min), prolonged PR interval, widening of QRS complex, long QT interval, T wave changes, ST-segment depression, atrial fibrillation, ventricular premature contractions (>three per minute), supraventricular arrhythmias, and ventricular arrhythmias.
The inclusion criteria were age groups from 18 to 65 yrs, major electively planned GI surgeries involving the gut lasting more than 3 h, and a minimum postoperative ICU stay of 72 h. The exclusion criteria were patients on loop diuretics or thiazides, recent magnesium supplementation, chronic alcoholism with withdrawal, and a past history of ischemic heart disease with ECG changes.
Apart from studying the incidence of hypomagnesemia and coexisting hypokalemia the statistical analysis was done using the following: Paired “t” test was used to test whether there was any significant difference between preoperative and postoperative serum magnesium and potassium values; Pearson correlation coefficient was used to assess the correlation between postoperative serum magnesium and potassium levels; chi-square test was used to test the association between low serum magnesium levels with ECG changes for that day. The data were analyzed using the SPSS 23.0 software.
Results
A total of 51 consecutive patients undergoing major gastrointestinal surgeries were studied, and these cases included conditions like colorectal malignancies, chronic pancreatitis, choledochal cyst, ulcerative colitis, duodenal adenocarcinomas, mesenteric gut ischemias, carcinoma stomach, oesophageal strictures, achalasia cardia, rectal prolapse, and subacute intestinal obstruction to name a few.
Twenty-eight (54.9%) patients were male, and twenty-three (45.1%) were female. The mean age of patients was 45.27 (SD = 16.28), with a minimum of 18 years and a maximum of 65 years. The study subjects included serving service personnel, retired personnel, their dependents, and also civilian patients admitted through the civil OPD, representing both urban and rural populations.
The mean values recorded for serum magnesium were 1.73 mg/dl (SD 0.32) in preop phase, 1.72 mg/dl (SD 0.22) on postop Day 2 and 1.71 mg/dl (SD 0.21) on postop Day 3 as depicted in Table 1. The mean values obtained for serum potassium were 4.03 meq/l, 4.14 meq/l, and 4.02 meq/l on a preop day, Day 2 postop, and Day 3 postop, respectively.
Table 1.
Range of serum Mg++ and K+ and their mean values on various days.
| Minimum | Maximum | Mean | Std. Deviation | |
|---|---|---|---|---|
| Preop Mg++ | 0.44 | 2.33 | 1.73 | 0.32 |
| Day 2 postop Mg++ | 1.17 | 2.31 | 1.72 | 0.22 |
| Day 3 postop Mg++ | 1.23 | 2.25 | 1.71 | 0.21 |
| Preop K+ | 3.10 | 5.90 | 4.03 | 0.62 |
| Day 2 postop K+ | 3.00 | 7.60 | 4.14 | 1.01 |
| Day 3 postop K+ | 2.70 | 7.60 | 4.02 | 0.83 |
N = 51 (Patients).
The primary objective was to study the incidence of hypomagnesemia in postoperative GI surgery patients. It was found that the incidence of preoperative hypomagnesemia was 43.1% with a 95% confidence interval (30.1–56.9) while postoperatively, it was 52.9% with a 95% confidence interval (39.2–66.2) on Day 2 and 47.1% with a 95% confidence interval (33.7–60.7) on Day 3, as depicted in Table 2.
Table 2.
Incidence of hypomagnesemia.
| Day of Mg++ Measurement | Hypomagnesemia | Frequency | Percent |
|---|---|---|---|
| Preoperatively | Absent | 29 | 56.9 |
| Present | 22 | 43.1 | |
| Total | N = 51 | 100.0 | |
| Day 2 postop | Absent | 24 | 47.1 |
| Present | 27 | 52.9 | |
| Total | N = 51 | 100.0 | |
| Day 3 postop | Absent | 27 | 52.9 |
| Present | 24 | 47.1 | |
| Total | N = 51 | 100.0 |
The incidence of preoperative hypokalemia was 17.6%, with a 95% confidence interval (8.9–29.9). On day 2, the incidence of hypokalemia was 27.5% with a 95% confidence interval (16.5–40.8), while on day 3, the incidence of hypokalemia was 19.6% with a 95% confidence interval (10.4–32.1), as depicted in Table 3.
Table 3.
Incidence of hypokalemia.
| Day of K+ measurement | Frequency | Percent | 95% confidence interval |
|---|---|---|---|
| Preoperatively | 9 | 17.6 | 8.9–29.9 |
| Day 2 postop | 14 | 27.5 | 16.5–40.8 |
| Day 3 postop | 10 | 19.6 | 10.4–32.1 |
Total (N = 51).
Any correlations between the serum magnesium and serum potassium levels on postop Days 2 and Day 3 were assessed using the Pearson Correlation Coefficient, as shown in Table 4. As depicted, the correlation between Day 2 postop serum magnesium (Mg++) and potassium (K+) levels is very poor, with a correlation coefficient of 0.23, which is statistically not significant (P = 0.11). Similarly, postop Day 3 Mg++ and postop Day 3 K+ show poor correlation (r = 0.28) with P = 0.04.
Table 4.
Correlations between Day 2 and 3 postop Mg++ and K+ levels.
| Day 2 Mg++ | Day 2 K+ | Day 3 Mg++ | Day 3 K+ | |
|---|---|---|---|---|
| Day 2 Mg++ | 1 | |||
| Day 2 K+ | 0.23 | 1 | ||
| Day 3 Mg++ | 0.78b | 0.15 | 1 | |
| Day 3 K+ | 0.28a | 0.66b | 0.28 | 1 |
Significant at 5% level of significance.
Significant at 1% level of significance.
However, postop Day 2 Mg++ shows statistically significant strong correlation with postop Day 3 Mg++ (r = 0.78, P = 0.001). As with serum Mg++, postop Day2 K+ shows statistically strong correlation with postop Day 3 K+ (r = 0.66, P = 0.001) as shown in Table 4. To test whether there was a significant difference between preop and postop serum magnesium and serum potassium values Paired “t” test was used. This is depicted in Table 5. No difference was seen in any parameter pre operatively and postoperatively.
Table 5.
Paired sample statistics.
| Paired Differences |
|||||
|---|---|---|---|---|---|
| Mean | Std. Deviation | 95% Confidence Interval of the Difference |
p value | ||
| Lower | Upper | ||||
| Pre-Op Mg++ - | |||||
| 0.01 | 0.29 | −0.06 | 0.09 | 0.72 | |
| Day 2 Mg++ | |||||
| Preop Mg++ - | |||||
| 0.02 | 0.32 | −0.06 | 0.11 | 0.63 | |
| Day 3 Mg++ | |||||
| Pre-Op K+ - | |||||
| 0.11 | 0.90 | −0.36 | 0.14 | 0.39 | |
| Day 2 K+ | |||||
| Preop K+ - | |||||
| 0.00 | 0.89 | −0.25 | 0.25 | 0.98 | |
| Day 3 K+ | |||||
As far as the secondary objective of the study was concerned i.e, to detect the presence of coexisting hypokalemia in these cases of postop hypomagnesemia, our study revealed that, on Day 2 postop, out of the 27 cases of hypomagnesemia, the incidence of hypokalemia was 33.3% and on Day 3 postop, out of the 24 cases of hypomagnesemia, the incidence of hypokalemia was 29.2% as depicted in Table 6.
Table 6.
Incidence of hypokalemia in cases of low serum Mg++ on Day 2 and 3 postop.
| Postop Day | Range | Frequency | Percent |
|---|---|---|---|
| Day 2 | High | 3 | 11.1 |
| Low | 9 | 33.3 | |
| Normal | 15 | 55.6 | |
| Total | 27 | 100.0 | |
| Day 3 | Low | 7 | 29.2 |
| Normal | 17 | 70.8 | |
| Total | 24 | 100.0 |
Overall, no significant difference was found between the incidence of postop hypokalemia in isolation when compared with the incidence of hypokalemia with coexisting hypomagnesemia both on Day 2 postop (27.5% vs 33.3%, P = 0.74) and on Day 3 postop (19.6% vs 29.2%, P = 0.16).
The other secondary objective of this study was to note any abnormal ECG patterns in these patients in the postop period. Any 12 lead ECG recorded on Day 2 and Day 3 postop was labeled as abnormal if it had any qualifying features as mentioned in material and method. None of the patients had an abnormal ECG preoperatively.
The incidence of abnormal ECG in cases with low serum Mg++ detected on Day 2 postop was 11/27 = 40.7% (95% confidence interval = 23.61–59.75) while on Day 3 postop it was 8/24 = 33.3% (95% confidence interval = 16.83–53.64).
Association of low serum Mg++ on Day 2 postop with ECG findings on that day was tested using the chi-square test. The analysis showed that there was a significant association between the incidence of hypomagnesemia on Day 2 and the ECG findings of that day (P = 0.02), as depicted in Table 7. This association was not found to be significant on the 3rd postop day.
Table 7.
Association of ECG with hypomagnesemia on Day 2 postop.
| ECG on Day 2 |
||||
|---|---|---|---|---|
| No change | Change | Total | ||
| Incidence of Hypomagnesemia | Absent | 21 | 3 | 24 |
| Present | 16 | 11 | 27 | |
| Total | 37 | 14 | 51 | |
P = 0.02 implying significant association.
As far as the incidence of abnormal ECG pattern in patients with coexisting hypomagnesemia and hypokalemia is concerned, it was found to be 33.3% on Day 2 postop, while on Day 3 postop, it was 28.6%, as shown in Table 8.
Table 8.
Incidence of abnormal ECG pattern in patients with coexisting hypomagnesemia and hypokalemia on Day 2 and 3 postop.
| Postop Day | Frequency | Percent | |
|---|---|---|---|
| Day 2 | No ECG change | 6 | 66.7 |
| ECG change | 3 | 33.3 | |
| N = 9 | 100.0 | ||
| Day 3 | No ECG change | 5 | 71.4 |
| ECG change | 2 | 28.6 | |
| N = 7 | 100.0 |
Overall, no significant difference was found between incidence of ECG abnormality in pure hypomagnesemia cases when compared to the incidence of abnormal ECG in cases of coexisting hypomagnesemia, as well as hypokalemia both on Day 2 postop (40.7% vs 33.3%, P = 0.75) and Day 3 postop (33.3% vs 28.6%, P = 0.90).
Discussion
Magnesium deficiency has been reported in 20%–65% of patients in ICU settings.11 Moreover, patients with hypomagnesemia in ICU exhibit mortality rates two to three times higher12 and prolonged hospitalization too.
In our study, the incidence of preop hypomagnesemia was 43.1% that further increased to 52.9% postoperatively on Day 2 and reduced to 47.1% on Day 3 (Table 2). Although this trend shows the presence of hypomagnesemia in nearly half the patient population afflicted with GI disorders who underwent a major GI surgery, however, no significant difference was found between preop and postop serum Mg++ values (Table 5) despite the postop Day 2 Mg++ levels showing a statistically significant strong correlation with postop Day 3 Mg++ levels (r = 0.78, P = 0.001).
Our values are comparable with other studies. In one study, serum magnesium (Mg++) concentration was measured in samples of blood collected from 193 patients admitted to postoperative ICUs wherein 117 patients (61%) had hypomagnesemia (serum Mg less than 1.5 mEq/L or 1.7 mg/dl). The study concluded that hypomagnesemia was common in postoperative ICU patients and patients with severe hypomagnesemia had a higher mortality.13 However, the type of surgeries were not mentioned in this study.
In another study, 171 cancer patients who underwent celiotomy procedures during a 38-month period were evaluated retrospectively for postoperative electrolyte alterations, especially serum magnesium.14 Hypomagnesemia was observed in those patients who had either undergone an operation with curative intent (P = 0.0035), a major resection (vs. no resection, P = 0.0259), or preoperative bowel cleansing with sodium phosphate.
Yet another study investigated the serum magnesium levels in 30 patients after major gastrointestinal surgery and evaluated if the magnesium-containing Plasma-Lyte 148 used as maintenance fluid influences these changes peri-operatively as compared against compound lactate solution. The procedures ranged from anterior rectal resection to thoracic-abdominal cardio-esophagectomy. The results showed a statistically significant (p < 0.05) reduction in magnesium levels in both groups.15 This showed that despite administration of magnesium-containing fluids, Mg++ levels remain low post major GI surgeries.
The reason for already preexisting hypomagnesemia in our patients preoperatively could be due to either the diseased pathology afflicting the small gut through which the prime absorption of Mg++ takes place or the bowel preparation given to the patients the night before surgery resulting in Mg++ loss through the gut. The levels recede further once a major GI surgery does take place, especially on Day 2 and Day 3 postop, as was evident in our study, although this drop was not statistically significant compared to the preop serum Mg++ values (Table 5). Studies on hypomagnesemia have documented that magnesium can be lost when intestinal secretions are incompletely reabsorbed, as with small-bowel bypass surgery.16 However, in the overall scheme of things, serum Mg++ levels become relevant for us since low levels could have far-reaching consequences on human physiology.
As far as serum potassium (K+) levels were concerned, the incidence of preop hypokalemia was 17.6%. On Day 2 postop, it was 27.5%, and on Day 3 postop, it was 19.6% (Table 3). Again, no statistically significant difference was found between preop and postop serum K+ values (Table 5). Also, similar to serum Mg++, postop Day 2 K+ showed statistically strong correlation with postop Day 3 K+ (r = 0.66, P = 0.001).
However, our secondary objective was not to simply detect hypokalemia in these cases but instead to detect the incidence of postop hypokalemia in cases, which had hypomagnesemia. The study revealed that on Day 2 postop, out of the 27 cases (52.9%) of hypomagnesemia, the incidence of hypokalemia was 33.3%, while on Day 3 postop, out of the 24 cases (47.1%) of hypomagnesemia, the incidence of hypokalemia was 29.2% (Table 6). This effectively makes 1/3rd of patients as having coexisting hypokalemia in cases with postop hypomagnesemia.
Hypokalemia is a common event in patients with hypomagnesemia, occurring in 40–60% of cases. In a recent study in patients undergoing gynecological oncology surgery in a tertiary care cancer institute, the incidence of hypokalemia with associated hypomagnesemia was 23%.17 This was relatively less as compared to the incidence of hypokalemia recorded with hypomagnesemia in our study. However, our cases were major GI surgery cases with gut involvement, and that could explain a higher incidence.
In a study by Whang et al on 46 hypokalemic patients with hypomagnesemia, it was concluded that hypokalemia could only be corrected when Mg++ depletion was corrected.18 Hence coexisting hypokalemia with hypomagnesemia assumes relevant proportions for our subset of patients.
The third objective of our study was to note any abnormal ECG patterns in these patients. 12 lead ECG recorded on Day 2 and Day 3 postop was labeled as abnormal if it had any qualifying features as mentioned in the material and method section in the form of unexplained tachycardia (Heart Rate > 120/min), prolonged PR interval, widening of QRS complex, long QT interval, T wave changes, ST-segment depression, atrial fibrillation, ventricular premature contractions (>3/min), supraventricular arrhythmias, or ventricular arrhythmias.
In our study, all the ECGs were analyzed but no major cardiac dysrhythmias were seen. The only abnormalities noted in 51 postop patients were significant sinus tachycardia in 04 cases, T wave inversions in 06 cases, and one case displaying the right bundle branch block pattern. However, while calculating the incidence of abnormal ECGs, these apparently innocuous changes, especially sinus tachycardia, were also factored in, keeping in mind the qualifying features we had set for defining an abnormal ECG.
Based on this, the incidence of abnormal ECG in cases with low serum Mg++ detected on Day 2 postop was 40.7% (95% confidence interval = 23.61–59.75) while on Day 3 postop it was 33.3% (95% confidence interval = 16.83–53.64). Association of low serum Mg++ on Day 2 postop with ECG findings on that day was tested using the chi-square test, and the analysis showed that there was a significant association between the incidence of hypomagnesemia on Day 2 with ECG findings of that day (P = 0.02) (Table 7). This association was not found significant on the 3rd postop day.
Since abnormal ECG patterns could also be seen with hypokalemia, we factored that in to note the incidence of abnormal ECG patterns in patients with coexisting hypomagnesemia and hypokalemia both. It was found to be 33.3% on Day 2 postop, while on Day 3 postop, it was found to be 28.6% (Table 8).
In one study, a prospective analysis of 62 patients admitted within 72 h after aneurysmal subarachnoid hemorrhage (SAH), an association of hypomagnesemia with several ECG abnormalities was dwelt upon.19 Of the patients, 61%were found to have a long QTc duration while low serum magnesium was related to a long PR interval (P = 0.001) and a shorter QTc interval (P = 0.004). In effect, in patients with SAH, low serum magnesium levels were found related to a less pronounced increase in the QTc interval. Our study found no change in QTc intervals whatsoever.
Sinus tachycardia and T wave inversions were present in a few patients along with postop hypomagnesemia in our study. Although sinus tachycardia could be explained as a part of postop pain syndrome or a postop inflammatory response/fever hence giving rise to a tendency to label it as non-significant considering that these GI surgeries were major in nature. But there are studies that have highlighted sinus tachycardia as the prime manifestation of Mg++ depletion followed by severe arrhythmia.20 Mg++ decreases intracellular calcium and acts as a calcium channel blocker.21 So lower levels of Mg++ would present with tachycardia as also negative T waves. Our study corroborated these findings. However, the few cases that did exhibit T wave inversions did not result in any major hemodynamic compromise. But, nevertheless, the association of abnormal ECG with hypomagnesemia was found strongest on Day 2 postop.
Conclusion
This study showed that hypomagnesemia was present in nearly half the patients, and there was a significant association between the incidence of hypomagnesemia and abnormal ECG on the second postoperative day, suggesting that serum magnesium be monitored postoperatively, and ECG evaluated routinely in postoperative hypomagnesemia. The limitation of the study was that it was time-bound, and the original sample size projected could not be achieved. A future study with a larger sample size would be able to expand on these findings.
Disclosure of competing interest
The authors have none to declare.
Acknowledgments
This paper is based on Armed Forces Medical Research Committee Project No. 4759/2016 granted by the office of the Directorate General Armed Forces Medical Services and Defence Research Development Organization, Government of India.
Authors also acknowledge Seema Patrikar, Lecturer in Statistics, Department of Community Medicine, Armed Forces Medical College, Pune.
References
- 1.Bringhurst F.R.D.M., Krane S.M., Kronenberg H.M. Harrison’s Principles of Internal Medicine. 18th ed. McGraw- Hill; New York: 2012. Bone and mineral metabolism in health and disease. [Google Scholar]
- 2.Vormann J. Magnesium and kidney health – more on the ‘forgotten electrolyte’. Am J Nephrol. 2016;44:379–380. doi: 10.1159/000450863. [DOI] [PubMed] [Google Scholar]
- 3.Hollifield J. Electrolyte disarray and cardiovascular disease. Am J Cardiol. 1989;63:21–26. doi: 10.1016/0002-9149(89)90934-x. [DOI] [PubMed] [Google Scholar]
- 4.Skoryna S.C., Edward D.W. Elsevier Science; 2014. Intestinal Absorption of Metal Ions, Trace Elements and Radionuclides. [Google Scholar]
- 5.Luis Noronha J., Matuschak George M. Magnesium in critical illness: metabolism, assessment, and treatment. Intensive Care Med. 2002;28:667–679. doi: 10.1007/s00134-002-1281-y. [DOI] [PubMed] [Google Scholar]
- 6.Whang R., Ryder K.W. Frequency of hypomagnesemia and hypermagnesemia: requested vs routine. J Am Med Assoc. 1990;263:3063–3064. [PubMed] [Google Scholar]
- 7.Chou-Long Huang, Kuo Elizabeth. Mechanism of hypokalemia in magnesium deficiency. J Am Soc Nephrol. 2007;18:2649–2652. doi: 10.1681/ASN.2007070792. [DOI] [PubMed] [Google Scholar]
- 8.Reinhart R.A. Magnesium metabolism: a review with special reference to the relationship between intracellular content and serum levels. Arch Intern Med. 1988;148:2415–2420. doi: 10.1001/archinte.148.11.2415. [DOI] [PubMed] [Google Scholar]
- 9.DiNicolantonio J.J., Liu J., O’ Keefe J.H. Magnesium for the prevention and treatment of cardiovascular disease. Open Heart. 2018;5 doi: 10.1136/openhrt-2018-000775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brusco L., Cohen S., Weissman C. Serum magnesium levels in abdominal and thoracic surgery. Crit Care Med. 1992;20:S19. [Google Scholar]
- 11.Jiang P., Lu Q., Lai T., Xu F. Does hypomagnesemia impact on the outcome of patients admitted to the intensive care unit? A systematic review and analysis. Shock. 2017;47:288–295. doi: 10.1097/SHK.0000000000000769. [DOI] [PubMed] [Google Scholar]
- 12.Rubeiz G.J., Thill-Baharozian M., Hardie D., Carlson R.W. Association of hypomagnesemia and mortality in acutely ill medical patients. Crit Care Med. 1993;21:203–209. doi: 10.1097/00003246-199302000-00010. [DOI] [PubMed] [Google Scholar]
- 13.Chernow B., Bamberger S., Stoiko M., et al. Hypomagnesemia in patients in postoperative intensive care. Chest. 1989;95(6):1362. doi: 10.1378/chest.95.2.391. [DOI] [PubMed] [Google Scholar]
- 14.Schwarz R.E., Nevarez K.Z. Hypomagnesemia after major abdominal operations in cancer patients: clinical implications. Arch Med Res. 2005 Jan-Feb;36(1):36–41. doi: 10.1016/j.arcmed.2005.01.001. [DOI] [PubMed] [Google Scholar]
- 15.Sanchez-Capuchino A., McConachie I. Peri-operative effect of major gastrointestinal surgery on serum magnesium. Anaesthesia. 1994;49:912–914. doi: 10.1111/j.1365-2044.1994.tb04275.x. [DOI] [PubMed] [Google Scholar]
- 16.Zalman S. Hypomagnesemia. JASN (J Am Soc Nephrol) 1999;10(7):1616–1622. doi: 10.1681/ASN.V1071616. [DOI] [PubMed] [Google Scholar]
- 17.Bhardwaj B., Desai A.D., Patel B.M., Parekh C.D., Patel S.M. Prevalence of hypomagnesemia in patients undergoing gynaecological oncology surgery in tertiary care cancer institute of India. Int J Reprod Contracept Obstet Gynecol. 2017 Oct;6(10):4355–4362. [Google Scholar]
- 18.Whang R. Clinical disorders of magnesium metabolism. Compr Ther. 1997;23:168–173. [PubMed] [Google Scholar]
- 19.Walter M., van den Bergh, Ale Algra. Electrocardiographic abnormalities and serum magnesium in patients with subarachnoid haemorrhage. Stroke. 2004;35:644–648. doi: 10.1161/01.STR.0000117092.38460.4F. [DOI] [PubMed] [Google Scholar]
- 20.Efstratiadis G., Sarigianni M., Gougourelas I. Hypomagnesemia and cardiovascular system. Hippokratia. 2006 Oct-Dec;10(4):147–152. [PMC free article] [PubMed] [Google Scholar]
- 21.Houston M. The role of magnesium in hypertension and cardiovascular disease. J Clin Hypertens. 2011;13:843–847. doi: 10.1111/j.1751-7176.2011.00538.x. [DOI] [PMC free article] [PubMed] [Google Scholar]