Interaction of Cobalamin Status With Nitrous Oxide in Relation to Postoperative Cognitive Changes in the Elderly

Subtle Disturbances of Cobalamin Status

Low cobalamin (vitamin B12) levels are frequent in the elderly. Most often they reflect a mild metabolic abnormality without clinical symptoms (subclinical cobalamin deficiency). It is unclear if these elderly people require medical intervention, unlike that small minority with clinical symptoms which can progress and create severe blood or nervous system problems. The study aims to determine if nitrous oxide (N2O), a common anesthetic agent, worsens cobalamin status in elderly patients with unrecognized subclinical cobalamin deficiency. The reason for concern is that N2O inactivates cobalamin and can aggravate the clinical picture of patients who already have clinical manifestations of cobalamin deficiency. The elderly are known to have an increased risk of developing mental changes after surgery and it may be that sometimes these result from aggravation of subclinical cobalamin deficiency.

The study recruits people over the age of 60 years who are undergoing clinically indicated elective surgery requiring general anesthesia for more than 1 hour. Patients meeting exclusion and inclusion criteria are randomized to receive either a standard anesthetic regimen that includes N2O or a nearly identical one without N2O. Before surgery and 2 weeks and 4 weeks after surgery, each patient undergoes (1) a broad battery of tests of cognition and mood and (2) blood tests measuring cobalamin, folate and homocysteine-methionine metabolism to determine whether they have any subtle biochemical impairment of cobalamin status. DNA from blood cells is also tested for the presence of common mutations that affect key enzymes in those metabolic pathways. A brief testing for postoperative delirium is also done 2 hours after surgery.

The patient subgroups' are analyzed for neuropsychologic changes over time, using the preoperative test as the baseline for all comparisons, and associations of those changes with metabolic, genetic, demographic and clinical data.

The primary question is what effect routine N2O exposure has on the latter compared with non-N2O anesthesia in elderly people who either have or do not have subclinical cobalamin deficiency. It will help answer whether or not the combination can help explain the increased risk of cognitive problems after surgery in elderly patients, and by extension whether preoperative cobalamin testing and treatment may be indicated in the elderly. It will also test whether genetic predisposition affects the described problems.

Study Design:

• Double-blind, randomized study of elderly patients scheduled for elective surgery under general anesthesia.
• Medical, cobalamin-related, and demographic information;
• Blood tests of cobalamin-related metabolic status and genotyping for 4 relevant mutations;
• Psychoneurologic evaluation before surgery;
• Randomization to receive a standard anesthetic regimen with nitrous oxide (N2O) or the same regimen without N2O;
• Intraoperative and perioperative data are collected;
• Testing for delirium 2-3 hours after surgery.
• All preoperative studies are repeated after 2 weeks and 4 weeks.
• Patients with abnormal cognitive status persisting at 4 weeks receive cobalamin and are retested at 3 months.

Inclusion criteria:

• Age >60 years;
• Anticipated duration of general anesthesia >1 hour;
• Good English comprehension.

Exclusion criteria:

• Surgery involving the brain, blood supply to head and neck, or heart;
• Dementia, psychosis or brain disease;
• Contraindication to N2O use;
• Clinical status other than ASA class 1 or 2;
• Bronchospastic or obstructive pulmonary disease;
• N2O exposure in past 6 months;
• Cobalamin injection in past 6 months;
• RBC mean corpuscular volume >95 fl;
• Creatinine >1.8 mg/dl.
• History or evidence of paresthesias, numbness of feet or hands, gait and balance problems, memory and thinking problems.

If the preoperative cognitive test produces a suspicious result (two or more unrelated test results >1.5 x standard deviation below the normal mean), immediate review is done to decide about exclusion of the patient from the study.

All investigators and testers are blinded to the N2O exposure (except the anesthesiologist) and cobalamin status.

Subject Numbers: Statistical power analysis by our statistician projected the need for 386 subjects to achieve a power of 80% with type I error of 0.05. Based on a possible 15% subject loss rate, 444 subjects are planned.

Patient Information questionnaire includes:

• Demographic information;
• Relevant medical history;
• Cobalamin-related history;
• Exclusion factors.

Randomization:

• Computer-generated numbers system;
• Made known just before surgery to the anesthesiologist only.

Anesthetic protocol:

• One of two effective, commonly used anesthetic regimens protocols;
• Standard ASA monitors, an oxygen analyzer and carbon dioxide capnograph are applied.
• General anesthesia is induced in both the N2O and non-N2O groups with fentanyl (2-5 µg/kg) and propofol (1-3 mg/kg), which allows for appropriate titration to the desired end point without adverse reactions. Cisatracurium (0.15 mg/kg) is given for muscle relaxation.
• Anesthesia is then maintained with either 40% oxygen/60% N2O (N2O study arm) or 40% oxygen in air (non-N2O arm), with desflurane for both study groups, titrated to depth of anesthesia. Fentanyl is used in both groups (1 µg/kg/hr) as the primary narcotic until just before the end of surgery;
• Following intubation, monitored ventilation maintains normocapnea and adequate oxygenation. Temperature, blood pressure, and heart rate are maintained by prescribed standard methods.
• At the end of surgery, desflurane and N2O (or, in the other study arm, desflurane and O2/room air) are stopped, residual neuromuscular block is pharmacologically antagonized, dolasetron is given as prophylaxis against nausea and vomiting, and postoperative analgesia is provided with opioids, as needed.
• The protocols allow for clinically indicated modifications at any time.
• All the above assures that anesthetic management is not compromised, allows the inclusion of N2O (vs room air) to be the only difference between the two arms, and uses no agents known to affect cobalamin other than N2O.

Intraoperative data collection:

• Duration of N2O exposure;
• Doses of all anesthetics;
• Surgery actually performed;
• Oxygen saturation, as monitored continuously;
• End-tidal CO2;
• End-tidal desflurane;
• Blood loss and transfusions;
• Temperature on emergence from anesthesia, and minimal temperature;
• Range of perioperative blood pressures;
• Presence of emergence delirium;
• Adverse postoperative events;
• Medications required postoperatively.

Blood Tests of Cobalamin Status and related tests:

• Serum cobalamin and folate levels;
• Homocysteine by immunoassay of EDTA-anticoagulated plasma separated within 1 hour in the PI's laboratory;
• Serum methylmalonic acid (MMA) by gas chromatography-mass spectrometry;
• Plasma metabolites (methionine, glutathione, S-adenosylmethionine, S-adenosylhomocysteine, cysteine, and cysteinylglycine; see later);
• Serum creatinine, complete blood count, and blood smear.

The diagnosis of subclinical cobalamin deficiency is made if BOTH of the following criteria are met:

• Two or more of the following 3 abnormalities: low cobalamin, high homocysteine, or high MMA levels. All investigators remain blinded to these findings, which require no action.
• Clinically apparent cobalamin deficiency is ruled out. This is always known by the investigators before surgery, permitting immediate exclusion of the patient from the study if positive.

Assay of Metabolites of Homocysteine and Methionine:

• Plasma is separated within 15 minutes of venipuncture, kept on ice until centrifugation, and stored in aliquots at -80C.
• Unthawed aliquots are shipped on dry ice to the University of Arkansas for assay using HPLC with coulometric detection.
• Because some metabolites are affected by renal status, adjustment for creatinine is made.
• Analyses focus on metabolite results in relation to (a) each other; (b) original cobalamin status; (c) folate levels; (d) N2O exposure, both in relation to original cobalamin status and gene polymorphisms; (e) presence of gene polymorphisms (including combinations); and (f) appearance of cognitive dysfunction vs none. Longitudinal metabolic changes over the entire follow-up period are examined.

Gene Polymorphism Analyses:

• DNA is extracted from preoperative blood samples, and purified DNA is stored at -80C.
• The 677C-T and 1298A-C mutations of MTHFR are determined by standard techniques;
• The methionine synthase 2756A-G (D919G) mutation is identified using HaeIII, and the MTRR 66A-G (I22M) mutation using NdeI.
• Analysis will consider homozygous mutation states and combined heterozygosity that produces impaired enzyme activity, combined mutations of the different enzymes, as well as allele frequencies.
• Study analyses focus on these questions: (a) Does metabolic or clinical N2O effect vary with genotype? (b) Do the polymorphisms influence the appearance of subclinical cobalamin deficiency? (c) How are metabolite changes affected by each mutation?

Cognitive Function and Depression Testing:

• Done by a trained tester in a quiet office free of distractions;

• A focused battery of neuropsychological tests that requires 1.5 hours is administered.

• The tests assess:

  • Attention/Concentration: (a) Paced Serial Addition Test, which requires active attention and rapid information processing; (b) California Computerized Assessment Package measures perfomance sensitive to attention and decision making.
  • Cognitive flexibility is tested by Letter-Number Sequencing, a subtest from the Wechsler Adult Intelligence Scale-III.
  • Memory and learning are tested with the California Verbal Learning Test-2.
  • Depression is assessed with the Beck Depression Inventory.

All tests are scored within 48 hours and reviewed with the neuropsychologist, both of whom are blinded to the patients' status. All measures are expected to show some "practice (familiarization) effect" over the 3 administrations, and appropriate adjustments and analyses are made.

If results in any test in two of the four test categories are >1.5 SD below normal mean in the preoperative assessment, immediate review for possible exclusion from the study is done. A decline in postoperative test results by the >1.5 x SD of the normal mean from the previous result in 2 individual tests is also brought to the attention of the PI and the safety monitor for a decision concerning cobalamin treatment.

Assessment for delirium is done 2-3 hours postoperatively with the Mini-Mental State Examination and the CAM-ICU test. Potential confounders such as drugs, hypoxia, and infection are taken into account in the analysis.

Long-term follow-up beyond 4 weeks is done in those patients with any cognitive function (including depression testing) significantly worse at the 4 week test than preoperatively. A decline in any test equivalent to 1 SD of the normal mean for that test is used as the decision benchmark.

Data Analysis:

• The primary question is whether patients with subclinical deficiency show neuropsychologic impairment after N2O exposure that is greater than in patients without deficiency. Analysis of these differences include comparison between deficient and non-deficient patients assigned to the non-N2O arm. Confounders are addressed by adjusting for relevant variables.
• The statistician conducts all analyses and consult on an ongoing basis throughout the study. Before analysis, continuous variables are subjected to logarithmic transformation. Univariate analyses, followed by selected multivariate analyses, include duration of anesthetic exposure, level of preoperative and postoperative metabolic deficits, as well as coexisting disorders. Bonferroni adjustment is used for multiple comparisons where appropriate. Multiple logitic regression is used to evaluate significant independent predictors of abnormal results.
• An interim analysis will be done after 3.5 years, using an alpha of 0.005.
• Analyses of metabolic and genetic findings will be done by univariate analyses, followed by multivariate analyses to determine the mutations' effects.

Risks to Subjects:

• All subjects are undergoing clinically indicated, elective surgery under general anesthesia. No aspect of their surgical management will be affected by their participation. The only study-related change in their clinical management will be the random assignment to omission or inclusion of N2O from near identical, standard anesthesia regimens.
• The only other manipulations are blood sampling and undergoing neuropsychologic testing several days before surgery and on 3-4 occasions postoperatively (except genetic tests, which are done only once).
• Medical information collection relevant to the study is subject to confidentiality protection in all HIPAA-mandated respects.
• Cobalamin/N2O related risks are actually smaller in participants than in nonparticipants. Several layers of protection against the small risks associated with the study include the cobalamin-related evaluation that is not otherwise done routinely before N2O anesthesia to identify patients who are at known risk from N2O; continued postoperative monitoring for changes in cobalamin status and significant cognitive decline; intervention with cobalamin therapy in possibly affected patients; institution of a Data Safety Monitoring plan with an independent safety monitor; and interim analysis of study findings.

Safety monitoring:

• Preoperative discovery of cognitive test dysfunction excludes patients from the study before any exposure to risk.
• A postoperative decline exceeding the equivalent of 1.5 x SD of the normal mean in two unrelated measures of cognitive function are brought to the immediate attention of the PI, the collaborating neuropsychologist, and an independent scientist-clinician who is not a part of our study.

Inclusion Criteria:

• age >60 years; elective surgery requiring general anesthesia for an hour or more; English speaker

Exclusion Criteria:

• Surgery involving CNS, blood supply to head, or cardiovascular system; preexisting dementia, psychosis or stroke; contraindication to any of the planned anesthetic agents; exposure to nitrous oxide in past 6 months; clinical status other than ASA class 1 or 2; bronchospastic or chronic obstructive lung disease; use of cobalamin injections in past 6 months; serum creatinine >1.8 mg.