Mannitol

Reducing raised intracranial pressure

The use of mannitol in the treatment of raised intracranial pressure has been reviewed, in the light of disagreements about the appropriate timing of administration, the optimal fluid management protocol, and the mechanisms of action of osmotic diuretics [ ].

The effects of four methods of infusion of mannitol and glycerol on raised intracranial pressure, as monitored by epidural pressure recordings, have been studied in 65 patients [ ].

• A.

  mannitol 0.5 g/kg was infused over 15, 30, or 60 minutes;

• B.

  mannitol 1.0 g/kg was infused over 30, 60, or 90 minutes;

• C.

  glycerol 0.5 g/kg in 5% fructose was infused over 30, 60, or 90 minutes;

• D.

  glycerol 1.0 g/kg was infused over 60, 120, or 180 minutes.

In group A, there were no differences in the reduction in intracranial pressure across the three infusion rates. In group B, the degree of reduction in intracranial pressure increased with shorter times of infusion. In groups C and D the reduction in intracranial pressure was inversely related to the rate of infusion. In each group, the slower the infusion rate of the same dosage, the longer the reduction in intracranial pressure lasted. There was a rebound increase in intracranial pressure in 12% of those given mannitol and 34% of those given glycerol. The dose and the rate of mannitol infusion did not affect the rebound.

In 22 patients with meningoencephalitis and hypertensive cranial syndrome from cerebral edema, mannitol was given to 13 and dexamethasone to nine [ ]. There were three therapeutic failures in those given mannitol and none in those given dexamethasone, although the two drugs had similar effects on the duration of the hypertensive cranial syndrome (39–44 hours). The patients who were treated with mannitol had hyponatremia after 48 hours.

In 43 patients, osmotherapy with mannitol 20% and sorbitol 40% increased the serum concentration of lactate but not pyruvate, causing an increased ratio of lactate to pyruvate [ ]. Sorbitol had the greater effect, with a maximum at 1 hour. Mannitol had its maximum effect at 4 hours. The author concluded that acidosis, shock, diabetes mellitus, and hepatic dysfunctions increase the risk of osmotherapy, especially with sorbitol.

Randomized trials of mannitol in patients with acute traumatic brain injury of any severity have been reviewed [ ]. In the pre-operative management of patients with acute intracranial hemorrhage high-dose mannitol reduced mortality (RR = 0.55; 95% CI = 0.36, 0.84) and reduced death and severe disability (RR = 0.58; 95% CI = 0.45, 0.74) compared with conventional-dose mannitol. In one trial treatment intended to lower intracranial pressure was compared with “standard care” (RR for death = 0.83; 95% CI = 0.47, 1.46). In one trial mannitol and pentobarbital were compared (RR for death = 0.85; 95% CI = 0.52, 1.38). In one trial pre-hospital mannitol was compared with placebo (RR for death = 1.75; 95% CI = 0.48, 6.38). The reviewers concluded that high-dose mannitol is preferable to conventional-dose mannitol in the pre-operative management of patients with acute intracranial hematomas. However, there is little evidence about the use of mannitol as a continuous infusion in patients with raised intracranial pressure who do not have an operable intracranial hematoma.

In 20 patients with head trauma and persistent coma who required infusions of an osmotic agent to treat episodes of intracranial hypertension resistant to standard modes of therapy, isovolumic infusions of either 7.5% hypertonic saline reduced the number of episodes of intracranial hypertension per day (6.9 versus 13.3) and the daily duration of episodes of intracranial hypertension (67 versus 131 minutes) compared with 20% mannitol [ ].

Stroke

In 805 patients who were given intravenous mannitol (mean dose, 47 g/day; mean duration, 6 days) or no treatment within 72 hours of the onset of a stroke, the case fatality was 25% versus 16% at 30 days and 38% versus 25% at 1 year [ ]. The prognostic scores on the Scandinavian Neurological Stroke Scale were similar in treated and untreated patients, both in ischemic and hemorrhagic strokes. However, the patient groups differed in several factors that might have affected survival. Thus, this uncontrolled study was inconclusive.

Serum and cerebrospinal fluid osmolarity were measured in 30 patients with severe head injuries or subarachnoid hemorrhage, 10 of whom received mannitol for at least 72 hours, 10 of whom received it for 24–48 hours, and 10 of whom were controls [ ]. Serum osmolarity increased quickly in all those who received mannitol and was unchanged in controls. Average cerebrospinal fluid osmolarity increased slowly in all those who received mannitol and was unchanged in controls. This is a potentially dangerous effect and the authors recommended that cerebrospinal fluid osmolarity should be measured regularly in all patients who receive mannitol for longer than 24 hours.

Disrupting the blood–brain barrier

Mannitol has been used to disrupt the blood–brain barrier temporarily in order to allow better penetration of chemotherapeutic drugs.

In eight patients with gliomas and one with a primary lymphoma of the central nervous system the blood–brain or blood-tumor barrier was reversibly opened by intra-arterial injection of hyperosmolar mannitol 25% [ ]. There was tumor regression or a tumor progression-free interval in five patients.

In 10 patients with malignant gliomas intra-arterial chemotherapy with 5-fluorouracil, nitrosourea, or interferon beta was given after osmotic blood–brain barrier disruption with intra-arterial 20% mannitol [ ]. In nine evaluable cases there were one complete and three partial responses; in five there was no change and no progressive disease on CT. The most untoward effect was myelosuppression: platelet and leukocyte counts fell below 20 × 10 ⁹ /l and 2 × 10 ⁹ /l respectively in three patients, of whom two died of severe infections. The other complications were eye pain during mannitol infusion in all cases in which selective catheterization of the internal carotid artery failed to pass the origin of the ophthalmic artery. There was reduced activity in 70%, nausea and vomiting in 50%, swelling of the external decompression area in 33%, and increased neurological deficits in 20%. However, all these adverse effects were transient.

Of 21 patients with malignant brain tumors, 16 were treated by operation, irradiation, and two or more courses of intracarotid infusion of nitrosourea 100 mg after 20% mannitol 200 ml, and five were treated similarly but without mannitol [ ]. The 2-year survival rate in those who received mannitol was 79% (11 of 14 cases followed for longer than 2 years) and the 3-year survival rate was 67%. Five of seven patients with grade 4 astrocytomas survived for more than 18 months, whereas four of five patients with grade 4 astrocytomas who did not receive mannitol died within 18 months.

Over 4 years, 37 patients with high-grade malignant gliomas underwent 246 treatment procedures with a combination of methotrexate, cyclophosphamide, and procarbazine given together with hyperosmolar mannitol-induced transient breakdown of the blood–brain barrier [ ]. There were complete remissions in 16% and 24 patients (65%) had partial or temporary remissions. Progression-free intervals were 1–47 (mean 15) months and median survival was 22 months. Neurotoxicity was minimal with one periprocedural death and five instances of worsened neurological deficits after a procedure.

The delivery of chemotherapeutic agents in the treatment of malignant brain tumors is improved by osmotic opening of the blood–brain barrier by prior infusion of mannitol into the internal carotid or vertebral artery. Over 4200 blood–brain barrier disruption procedures have been performed in over 400 patients with primary central nervous system lymphomas, gliomas, primitive neuroectodermal tumors, germ cell and metastatic cancers in the National Blood–brain Barrier Program [ ]. In patients with primary nervous system lymphomas, long-lasting responses have been obtained without loss of cognitive function and without the use of radiotherapy. The results in patients with primitive neuroectodermal tumors and germ cell tumors are also said to be very encouraging.

The efficacy of mannitol in augmenting the tumoricidal effect of etoposide has been studied in 99 children aged 1–21 years with recurrent brain tumors [ ]. They were randomly assigned to intravenous etoposide 150 mg/m ² with or without mannitol 15 g/m ² , daily for 5 days every 3 weeks for 1 year or until disease progression or death. CT or MRI scans, obtained after three cycles of therapy, were compared with pre-therapy scans. Of 87 evaluable patients, 12 had an objective response according to the radiologist and of 66 patients reviewed centrally, seven responded (two of 12 low grade astrocytomas, four of 26 medulloblastomas or primitive neuroectodermal tumors, one of 13 high-grade astrocytomas, and one of 15 brain stem gliomas). Survival at 1 year was 53% for low grade astrocytomas, 38% for medulloblastomas or primitive neuroectodermal tumors, 28% for high-grade astrocytomas and 9% for brain stem gliomas. Mannitol had no beneficial effect.

Use in bowel cleansing

Mannitol has been used for preoperative bowel cleansing before radiological investigations [ ], diagnostic and operative endoscopy [ ], and bowel surgery.

In whole gut irrigation mannitol is badly tolerated and leaves a bowel full of gas and fluid although it causes only small changes in serum electrolytes [ ]. In a study of the effect of an intravenous infusion of saline on the volume of rectal effluent and quality of bowel preparation produced by a smaller oral dose of mannitol, 19 patients drank 2–3 l of 5% mannitol, supplemented by an intravenous infusion of isotonic saline and 19 patients drank 4–5 l of 5% mannitol [ ]. The volume of rectal effluent and the quality of bowel preparation was the same in both groups. Loss of sodium in the oral group was corrected by the intravenous infusion, but the infusion resulted in greater water retention. There was no difference in the incidence of vomiting between the two groups.

Polyethylene glycol electrolyte lavage solution has been compared with 10% mannitol for preoperative colonic cleansing in 80 patients [ ]. Colonic cleansing was better with polyethylene glycol (90% optimal cleansing versus 75%). Mannitol caused subclinical dehydration according to hematological, biochemical, and weight changes before and after bowel preparation and caused more nausea, cramps, and abdominal pain. Two patients given mannitol had combustible amounts of hydrogen gas in the colon.

Three formulations, two based on magnesium citrate and one an optimized oral mannitol regimen, have been compared for their effectiveness in clearing the large bowel before double-contrast barium enema and for effects on barium mucosal coating [ ]. The formulations based on magnesium citrate were equally good and caused significantly less nausea and vomiting than mannitol. The authors concluded that mannitol should not be used for preparing the bowel for barium enema.

Two hypotonic non-hemolysing irrigating solutions, sorbitol + mannitol (2% + 1%) and glycine (1.5%), have been compared in patients undergoing transurethral resection of the prostate [ ]. Ethanol (1%) was added to the irrigating fluid as a marker to allow early detection of fluid absorption by breath analysis. There was very little absorption (less than 1 liter).

However, in other cases large volumes of fluid have been absorbed. In 39 patients having transurethral resection of the prostate for benign prostatic hyperplasia, large quantities of mannitol, which was used as the irrigating fluid, entered the circulation [ ]. There was a corresponding fall in serum sodium concentration. Patients who had serum mannitol concentrations over 4 mg/ml had hypotension and bradycardia; because they were nearly all hypovolemic, the bradycardia was thought to be inappropriate.

Irrigating fluid bags containing mannitol 3% or glycine 1.5%, both with added ethanol 1% as an indicator of fluid absorption, were used to investigate adverse effects in a randomized, double-blind study during 394 transurethral prostatic resections [ ]. The incidence of 13 symptoms was studied in 52 patients (13%) who absorbed more than 500 ml of fluid. The incidence of circulatory symptoms did not differ between the fluids, but the risk of neurological symptoms, such as nausea, was 4.8 times higher with glycine 1.5%. An increase of 1000 ml in the volume of irrigant absorbed increased the overall risk of circulatory symptoms by a factor of 3.4 and the risk of neurological symptoms by a factor of 4.4. The authors concluded that absorption of mannitol 3% during transurethral prostatic resection is associated with fewer neurological symptoms than glycine 1.5%.

In 80 patients randomized for precolonoscopic cleansing with either 10% mannitol 750 ml or sodium phosphate 180 ml there were statistically significant differences in serum sodium, phosphorus, potassium, and calcium between the two groups, but no clinical symptoms and no significant differences in the frequencies of adverse effects [ ]. Six of eight patients who were treated with sodium phosphate and who had had mannitol for a previous colonoscopy preferred sodium phosphate. The endoscopists, who were blinded to the treatment, reported excellent or good bowel preparation in 85% of those prepared with sodium phosphate versus 83% for mannitol. The authors concluded that although the quality of preparation and the frequencies of adverse effects were similar with the two solutions, retention of sodium and phosphate ions contraindicates the use of sodium phosphate in patients with renal insufficiency, cirrhosis, ascites, and heart failure.

In a retrospective study of patients who underwent elective surgery for colorectal carcinomas, traditional bowel preparation was performed the day before the operation either with oral castor oil 30 ml and three soap enemas (n = 154) or with mannitol 500 ml (n = 36) [ ]. There were infectious wound complications in 26 patients (17%) pretreated with castor oil compared with 13 patients (36%) treated with mannitol. There were no differences in the incidence of anastomotic leaks or mortality rate.

Systemic antimicrobial prophylaxis with metronidazole and gentamicin has been compared with metronidazole alone in elective colorectal surgery in a prospective randomized trial, in which all the patients received 10% mannitol solution before surgery [ ]. Although there were no serious infections in either group, the incidence of superficial wound infections was relatively high: 19% in those given metronidazole and gentamicin prophylaxis and 25% in those given metronidazole alone. Escherichia coli was isolated from all these wounds, and no obligate anerobic bacteria were cultured. The high rate of wound infection was probably caused by overgrowth after irrigation, due to residues of mannitol in the colon, which serve as a nutrient for Escherichia coli .

Both sets of authors concluded that mannitol should not be used for preoperative mechanical preparation of the large bowel before elective colorectal surgery.

Hydrogen gas can accumulate in the colon after the administration of mannitol [ ]. This can cause a risk of explosion.

Colonic explosion during colonoscopic polypectomy occurred after mannitol had been used for bowel preparation and the colon was completely clean [ ]. In spite of emergency surgery, with transfusion of 45 units of blood, uncontrollable hemorrhage persisted from multiple bleeding points and the patient died.

Use as a radiocontrast medium

In 56 patients undergoing abdominal CT the gastrointestinal tract was defined by negative contrast with 2.5% mannitol instead of the conventional positive contrast from an iodine-containing contrast medium [ ]. The number of artifacts due to high-contrast boundaries was slightly greater with negative contrast than it would have been with positive contrast, but differentiation of the gastrointestinal tract from other abdominal organs was equally good. Negative contrast was poor for diagnosing cystic tumors but much better than positive contrast for evaluating the wall of the gastrointestinal tract.

The effect of oral mannitol in an aqueous solution in enhancing pelvic MRI has been reported in a retrospective study in 72 patients with suspected or proven pelvic abnormalities: In 36 patients bowel marking was not carried out and in 36 patients the bowel was contrast-enhanced by oral mannitol 1000 ml [ ]. Mannitol significantly improved delineation of the intestinal structures and pelvic organs or pathological lesions, but eight patients had diarrhea, nausea, or meteorism.