Marieke Smelt

Bariatric surgery patients have a higher risk of developing nutritional deficiencies due to the anatomical alterations of the gastro-intestinal tract. Vitamin deficiencies is one of the most frequent problems within majority of bariatric patients. Therefore, prevention, detection and treatment of vitamin deficiencies are key aspects of long-term nutritional surveillance. The general aim of this thesis was to optimize all aspects regarding vitamin deficiencies after bariatric and metabolic surgery, of which patient adherence is the most important one. To achieve this aim, six research questions were defined to assess several important issues encountered in daily clinical practice:

1. Can we reduce the percentage of vitamin deficiencies by using the specialized weight loss surgery (WLS) multivitamin supplement (MVS)?
2. How can we improve the detection of vitamin B12 deficiencies after bariatric surgery?
3. What is the biochemical effect of different vitamin B12 supplementation regimens in deficient patients?
4. Does vitamin B12 treatment in patient with serum vitamin B12 between 140 and 200 pmol/L result in biochemical and clinical improvement?
5. What is the effect of different cholecalciferol supplementation regimens on biochemical outcomes and physical fitness?
6. Which factors and facilitators affect patient adherence to MVS intake?

Part I: Nutritional assessment in bariatric and metabolic surgery

In Chapter 2 we compared the effectiveness of specialized WLS MVS compared with a regular MVS after sleeve gastrectomy (SG) in 970 patients. The use of WLS MVS resulted in significantly higher mean serum vitamin B1, vitamin D, vitamin B12, folic acid, and ferritin. WLS MVS resulted in less vitamin B1, vitamin B12 and folic acid deficiencies. Anemia, iron deficiencies and hypervitaminosis B6 were diagnosed more often in WLS MVS users, suggesting that the iron dose in this supplement is insufficient and vitamin B6 dosage too high. The total number of de novo deficiencies was significantly reduced during 4 years for all WLS-users. This shows that SG patients benefit from these specialized supplements. However, adjustments of WLS formulas are required for vitamin B6 and iron. This study shows the importance of lifelong follow-up, because WLS MVS cannot prevent all vitamin deficiencies.

In Chapter 3, we reviewed the biochemical effect of preventive vitamin B12 supplementation and different vitamin B12 treatment regimens in deficient patients. Ten studies were included in this systematic review. The dosage of preventive oral vitamin B12 supplementation in all included studies varied from 1 µg/day to 600 µg/day. Eight of the included studies described significant persistence of vitamin B12 deficiencies when taking oral vitamin B12 doses below 350 µg/day. Therefore, this review suggests, that the oral dosage of vitamin B12 should be at least 350 µg/day in order maintain normal serum vitamin B12 concentrations. The influence of patient adherence to MVS use was only reported in two studies and remains therefore unclear.

In Chapter 4, we analyzed whether intramuscular (im) hydroxocobalamin injections in patients with serum vitamin B12 between 140 and 200 pmol/L results in improvement of serum vitamin B12, percentage of deficiencies and clinical symptoms. A vitamin B12 deficiency was defined as serum methyl malonic acid (MMA) > 300 nmol/L. All vitamin B12 deficiencies were treated successfully, and serum MMA normalized in all patients of the treatment group. Reported complaints in the intervention group (29%), such as tingling fingers, reduced attention span and tiredness, resolved after treatment. The control group, which did not receive im injections, showed a significant increase in serum MMA. In addition to this, the deficiency rate increased from 13% to 53% within six months, fortunately without increase of symptoms. This study shows that all vitamin B12 deficient patients benefit biochemically and clinically from im hydroxocobalamin treatment. However, beneficial effects were also found in patients without biochemical abnormalities, but with vitamin B12 related complaints. Patients with normal serum vitamin B12 values may have subclinical elevated serum MMA values and thus benefit from treatment. Therefore, diagnosis with combined parameters (serum vitamin B12 and MMA) is important. This study suggests that im treatment itself results in clinical improvement in patients with serum vitamin B12 between 140 and 200 pmol/L. With the results of this study an algorithm was developed and implemented at the obesity center of the Catharina Hospital Eindhoven (Figure 1).

Figure 1: Algorithm for detecting vitamin B12 deficiency
Serum vitamin B12: < 200 pmol/L: Deficient; > 200 - < 300 pmol/L: Check Serum MMA; > 300 pmol/L: Not deficient.
Serum MMA: > 430 nmol/L: Severe deficient; > 300 - < 430 nmol/L: Moderate deficient; < 300 nmol/L: Not deficient. In Chapter 5, we aimed to analyze the biochemical effects of different im hydroxocobalamin treatment regimen in a retrospective matched cohort study. Three regimens were compared in patients with a moderate to severe vitamin B12 deficiency based on serum MMA (Figure 1): three im injections, six im injections and a control group without im injections. Treatment with six im injections resulted in biochemical normalization, a faster decline of serum MMA and clinical recovery in all patients. Treatment with three im injections led to persistent deficiencies in 38% of the patients. All patients in the control group showed an increase of MMA without spontaneous improvements. In conclusion, this study suggests that a regimen with six im injections is sufficient to treat all vitamin B12 deficiencies biochemically. In Chapter 6, we aimed to analyze the biochemical effects of two different cholecalciferol treatment regimens to prevent a vitamin D deficiency in a retrospective matched cohort study. Group A received 800 IU cholecalciferol daily and group B received 800 IU cholecalciferol daily and 50,000 IU cholecalciferol once per month. Serum 25-hydroxyvitamin D increased from 37.8 ± 20.6 nmol/L to 66.7 ± 18.5 nmol/L in group A versus 47.0 ± 21.5 nmol/L to 94.2 ± 25.7 nmol/L in group B. Persistent vitamin D deficiencies were observed in 70% and 20% after six months for group A and B, respectively. In conclusion, a cholecalciferol regimen with 800 IU daily and 50,000 IU monthly leads to a faster increase of serum 25-hydroxyvitamin D and less vitamin D deficiencies. However, this cholecalciferol regimen needs to be further optimized, as it is still not sufficient to treat all vitamin D deficiencies. In our second vitamin D study (Chapter 7), the same database and cholecalciferol treatment regimens were used. We assessed the effect of the two cholecalciferol treatment regimens and protein intake on physical fitness in a retrospective matched cohort study. The protein intake was calculated by the dietician with a 24-hour food intake registration form three and six months after surgery. Outcomes of the handgrip strength (HS) and shuttle walk run test (SWRT) were used to analyze the effect on different supplementation regimens on physical fitness three and six months postoperatively. Outcomes of these two tests were assessed by the physiotherapist. Results showed that HS and SWRT were not significantly influenced by serum 25-hydroxyvitamin D. Multivariate analysis showed that protein intake influences postoperative HS outcome measurements only. Part II: Adherence to multivitamin supplementation after bariatric and metabolic surgery Chapter 8 attempted to clarify which potential barriers and facilitators influence patient adherence to MVS use after bariatric surgery. Because there is only limited data on this subject in bariatric patients, we gained knowledge from other patient populations with chronic medication or supplementation usage in order to find potential factors that can improve patient adherence in the obese population. Table 1 gives an overview of the negative influencing factors of this narrative review. These results were used to develop a questionnaire to assess patient adherence to MVS intake and served as a preliminary study for Chapter 9. Table 1: Overview of this review of barriers who negatively influences patient adherence. Patient-related factors: Bariatric patients: Age, Gender, Employment, Postoperative complications, Postoperative complaints, Eating behavior. Other patient populations: Age, Low education levels and income. Therapy-related factors: Bariatric patients: Forgetting, Swallowing, Disintegration properties of MVS. Other patient populations: Long duration of the medical treatment, High frequency of daily dose, Convenient way of administration, Adverse effects of medication. Psychosocial and economic factors: Bariatric patients: High costs of MVS treatment. Other patient populations: Patients lack belief in benefit of treatment, Erroneous beliefs or misconception, Negative attitude toward medication, Treatment of asymptomatic disease, Presence of psychological problems, Forgetfulness or other priorities, None or limited emotional support, High costs of medical therapy. Healthcare-related factors: Bariatric patients: Cancelling yearly medical visits. Other patient populations: Low satisfaction with clinical visits, Inadequate follow-up due to missing appointments, Discharge planning, Poor patient-doctor relationship. In Chapter 9 we described the results of a cross-sectional multicenter survey study about patient adherence to MVS use after bariatric surgery. Four high-volume obesity centers in the Netherlands participated in the study and 15,424 patients were recruited. In total, 4614 patients (29.9%) were included in this study. Self-reported anonymized answers from MVS-users (n = 4274, 92.6%) were compared with those from the non-users (n = 340, 7.4%). Of all MVS users, 15.4% did not take MVS regularly. Most frequent reasons were ‘forgetting daily intake’ (68.3%), ‘gastro-intestinal side effects’ (25.6%) and ‘unpleasant taste or smell’ (22.7%). For non-users, the gastro-intestinal side effects after MVS intake (58.5%), costs (13.5%) and absence of deficiencies (20.9%) were the most frequent reasons to stop taking MVS. Most important differences between MVS uses and non-users were the occurrence of gastro-intestinal complaints (37.4% vs 26.3%) and the frequency of occurring. Non-users experienced post-prandial complaints multiple times a week or day whereas MVS users experienced these complaints less frequently. Half of all included patients indicated that these complaints were not discussed by the healthcare professional during medical consultation. Other significant differences between non-users and MVS users were costs of the supplements (61.2% vs. 39.4%) and many patients think that adherence to their MVS intake could improve by reimbursement of the supplements (43.5% vs. 38.1%). Non-users were more often dissatisfied about instructions provided about MVS use. Attention paid to MVS use during medical consultations and the extent to which personal preferences of MVS use were also taken into account. Most frequent reasons for all included patients to score poor or inadequate (n = 1315, 28.5%) on the satisfaction scale was ‘too general information’ (57.1%), ‘personal preferences are not taken into account’ (51.0%) and ‘not enough time for adequate information about MVS during medical consultation’ (36.5%). Other reasons reported by patients were: ‘too short consultation time’ (23.5%), ‘having to actively ask for information themselves’ (28.9%), ‘only being told what he/she is doing wrong’ (9.4%) and ‘only one MVS formula is advised by the healthcare professional’ (6.5%). In conclusion, the attitude of many bariatric patients towards MVS use is predominantly negative. This suggests it is important to provide more extensive information about different possibilities in MVS use. The advantages and disadvantages of MVS should be provided. In addition to this, beliefs and thoughts of the patient should be obtained in order to establish the patient’s personal preferences. Challenges lie in improving patient adherence by implementing a shared decision-making process about MVS use, further optimization of the tolerance of WLS MVS formulas and exploring options for reimbursement of these specialized WLS MVS. PART III