1 INTRODUCTION

Obesity is a global health issue today with worldwide rates tripling since 1975. In 2016, 650 million adults (13%) were reported as obese that is, a Body Mass Index (BMI) greater than or equal to 30 kg/m2 (WHO, 2020). The corresponding proportion in Sweden is 15%, according to a recent national health survey (The Public Health Agency of Sweden, 2018). Obesity is considered a risk factor for an array of chronic diseases (The Global Burden of Disease 2015 Collaborators, 2017) however, studies on the association between an obese condition and oral disease show somewhat contradictory results and causal relationships are not well established (Nascimento et al., 2016; Shivakumar, Srivastava, & Shivakumar, 2018).

Obesity is difficult to treat despite several treatment options, including reduced dietary energy intake, physical activity, pharmaceutical treatment, and surgery (Bray, Fruhbeck, Ryan, & Wilding, 2016). Bariatric surgery is the most effective treatment for morbid obesity that is a BMI≥40 kg/m2 or a BMI≥35 kg/m2 in presence of obesity-related comorbidities f.i. diabetes or hypertension (NIH conference, 1991), and results in effective weight loss and reduced comorbidities (O'Brien et al., 2019). The number of surgeries has increased steadily in recent decades with gastric bypass surgery (GBP) and sleeve gastrectomy (SG) being the most frequently used surgical methods today (Angrisani et al., 2018). GBP involves formation of a small pouch from the upper part of the stomach and connecting it to the small intestine, enabling ingested food to pass directly into the intestinal tract. With SG, the greater part (3/4) of the stomach is removed and the rest of the stomach is formed into a narrow tube. Before qualifying for surgery, subjects need to reduce their body weight 5–10% through dietary modifications, to minimize intra- and postoperative complications (Anderin, Gustafsson, Heijbel, & Thorell, 2015).

Despite the many benefits of bariatric surgery, the restructuring of the gastrointestinal tract may cause complications. The short-term complications are mainly due to the surgical procedure, while long-term complications are the consequence of the restrictive and malabsorptive process that occurs after bariatric surgery. Known long-term complications are i.a. abdominal pain, gastroesophageal reflux disease (GERD) and nutritional deficiencies (Schulman et al., 2017). However, the long-term adverse effects are not fully elucidated (Höskuldsdottir et al., 2020).

The impact of bariatric surgery on oral health is sparsely studied and the findings contradictory. Some studies indicate associations between bariatric surgery and periodontal disease (Marsicano, Sales-Peres, Ceneviva, & Sales-Peres, 2012; Sales-Peres et al., 2015), dental caries (Hague & Baechle, 2008; Marsicano, Grec, Belarmino, Ceneviva, & Peres, 2011; Salgado-Peralvo et al., 2018) and tooth hypersensitivity (Netto et al., 2012), while others did not (Cardozo et al., 2014; Jaiswal et al., 2015). Biomedical characteristics and oral behaviors after surgery need for example to be explored. The studies have often focused on a single problem f.i. dental caries; however, there is no overall picture of possible oral problems facing the post-bariatric individual. Hence, knowledge in the area is fragmented.

Taken together, the oral health problems in post-bariatric individuals need to be thoroughly described in order to generate hypotheses for longitudinal studies in the field. Thus, the aim of the present study was to describe the oral health profile of individuals who had undergone the most common bariatric surgery procedures, gastric bypass or sleeve gastrectomy.

2 MATERIAL AND METHODS

A case series was designed (A Dictionary of Epidemiology, 2014). The inclusion criteria were having undergone GBP or SG surgery (Angrisani et al., 2018) ≥2 years earlier, together with professionally observed and/or patient-perceived aggravated oral problems after surgery. The exclusion criteria were treatment with other bariatric surgeries than GBP or SG or individuals who had their original bariatric procedure rearranged to GBP or SG. The surgical technique used was verified by medical records. After surgery, all had been asked to adhere to standardized daily nutritional substitution adapted to age and gender according to guidelines (Laurenius, Näslund, Sandvik, Videhult, & Wirén, 2018).

The Regional Ethical Review Board in Gothenburg (reg. no. 237-16) approved the study and all participants provided written consent.

2.1 Procedure and variables

Recruitment took place through a network of dental practitioners and health care professionals, such as physicians, dieticians and nurses in the western part of Sweden, who in their work identified patients with oral health problems after bariatric surgery. This selection generated a kind of convenience sample.

The participants filled out a 4-day food record prior to the dental visit. They were instructed to eat according to their normal everyday meal patterns while keeping a detailed record during four consecutive days (three weekdays and one weekend day). Instructions were also given to express the quantities using common household measures and kitchen scales, and to provide detailed descriptions of the food items, portion sizes and type of meals. The food records were analyzed by a registered dietician (SB) using the software Dietist Net Pro® and the National Food Composition Tables version 20171215. Meal frequencies and total daily energy intake were calculated and the distribution of the nutrient content was analyzed.

At the dental visit, the participants first filled in a questionnaire. Oral health habits were represented by tooth brushing (twice a day, vs. once a day/a few times a week/seldom/never), interdental cleaning (everyday/3–5 times a week, vs. 1–2 times a week/never), frequency of dental appointments during the last 5 years (regularly/at least once a year/3–4 appointments, vs. 1–2 appointments/never), and the reason for the last dental visit (routine examination, vs. pain/other problem). Possible oral problems included tooth hypersensitivity, acid reflux episodes and vomiting episodes (never/a few times a year, vs. a few times a month/a few times a week/daily). The participants rated their chewing ability (without difficulty, vs. difficult/unable). Their Oral Health-Related Quality of Life (OHRQoL) was assessed using the Oral Health Impact Profile (OHIP-14) (Slade, 1997). The response options for each item were “never” (0), “hardly ever” (1), “occasionally” (2), “fairly often” (3) and “very often” (4) with a possible maximum of 56, with a higher score indicating a greater impact.

A medical history was recorded in connection with the dental examination. Height (cm) and weight (kg) wearing light clothes were registered and Body Mass Index (BMI) was calculated (kg/m2).

Thereafter, unstimulated salivary secretion rate was measured for 15 minutes. Stimulated saliva was collected for 5 min with the participant actively chewing a paraffin pellet and spitting continuously. The salivary buffer capacity (low/medium/high) of the stimulated saliva was determined chairside (CRT Buffer®). Total bacterial count, total streptococci count, streptococcus mutans (SM) and lactobacilli, all per colony-forming unit (CFU)/ml of stimulated saliva were analyzed at the laboratory of the Department of Cariology, Institute of Odontology, University of Gothenburg, Sweden. Salivary electrolytes were analyzed at the Clinical Chemistry laboratory, Sahlgrenska University Hospital, Gothenburg, Sweden (Klingberg et al., 2007).

The clinical oral examination was then performed by one of two calibrated dentists (NT and ALÖ). The third molars were excluded from all examinations.

Dental caries was registered by visual and radiographic examination. Visual assessments of all tooth surfaces (buccal, lingual, distal, mesial, occlusal) were made according to the International Caries Assessment System II (ICDAS) (Ismail et al., 2007) with scores from 0 (sound surface) to 6 (distinct excessive cavity with visible dentine). Restorations and missing teeth were noted. The radiographic assessment was performed on four bitewing radiographs according to ICDAS, on three surfaces of the molars and premolars (distal, mesial, occlusal), and scored from 0 (no radiolucency) to 6 (radiolucency into the pulp). Detailed coding schemes are attached as Table S1. Clinical and radiographic assessments were weighed together for all surfaces recorded on radiographs. When the deviation between the clinical and the radiological assessment was maximum one step, the higher recorded value was used. When the difference was greater, the radiographs were reexamined (number of surfaces = 36). All these differences occurred when caries was diagnosed on radiographs under old fillings; that is, difficult to detect clinically.

Periodontal status was represented by (a) visible dental plaque, 0/1; (b) gingivitis, bleeding on probing, (0/1) and (c) probing pocket depths (measured from the gingival margin): healthy (<3.5 mm) or pathological (≥ 3.5 mm) (Löe, 1967; Ramfjord, 1967).

2.2 Statistical analysis

Data management and analyses were carried out using SPSS version 25. Results are presented with descriptive statistics including mean values, SDs and minimum-maximum scores.

3 RESULTS

A total of 14 subjects were examined, with an age range of 31–66 years. Six participants were younger than 40 years, five were between 40 and 59 years old and three were 60 years or older. Two were males. All participants had a BMI > 25 (range 25.4–44.7) at the time of the clinical examination. They had undergone bariatric surgery 2–19 years ago (mean 13 years) (Table 1). The majority had been treated with GBP and one with SG (case 14). All reported adhering to the standardized substitution therapy recommended after bariatric surgery (Höskuldsdottir et al., 2020; Laurenius et al., 2018). Six of the participants stated regular use of prescription medication, such as painkillers and antidepressants. All had a minimum of compulsory education (9 years) and all but four were cohabiting (not in tables).

TABLE 1. Description of the sample Case Age span Gender Surgical technique Years since operation Included via Main experienced oral problem after surgery BMI at clinical examination after surgery 1 40–59 Female GBP >2 yearsa Dietician Fractures, brittle teeth 38.9 2 21–39 Male GBP 10 Dietician Tooth hypersensitivity 26.0 3 21–39 Female GBP 9 Physician Tooth hypersensitivity a 4 21–39 Female GBP 11 Physician Bleeding gums 39.5 5 60–66 Female GBP 8 Dietician Xerostomia 40.4 6 60–66 Female GBP 13 Dietician Tooth decay, periodontitis 30.3 7 40–59 Female GBP 13 Dietician Tooth decay 44.7 8 40–59 Female GBP 19 Dentist Tooth decay, brittle teeth 25.9 9 40–59 Female GBP 11 Dentist Periodontitis 31.6 10 60–66 Female GBP 19 Dentist Tooth decay, brittle teeth 34.7 11 40–59 Female GBP 9 Dentist Tooth decay 25.4 12 21–39 Female GBP 9 Dentist Tooth decay, brittle teeth 32.5 13 21–39 Female GBP 12 Nurse Tooth decay, brittle teeth 35.4 14 21–39 Male SG 2 Dentist Tooth decay 27.8

The number of remaining teeth ranged between 16 and 28 (mean 24), with only four participants being fully dentate. The proportion of clinically registered dental caries of any stage, ICDAS codes 1-6, on the total number of examined surfaces in all participants was 26% (not in tables). Two-thirds of all lesions were categorized as initial non-cavitated caries (ICDAS 1-2).

Dental caries and periodontal status, for all 14 participants separately, are presented in Table 2. All participants exhibited dental caries with the number of decayed surfaces ranging between 7 and 80. Eleven of the fourteen participants showed severely decayed surfaces (ICDAS 5-6). Two participants had nine and one participant had ten such surfaces. All participants had previous restorations (filled surfaces mean 29.9%, range 2.1–90.8%). The oral hygiene was poor (visible plaque mean 59.2%) and bleeding on probing frequent (mean 31.1%). Pathological periodontal pockets were less frequent. However, one of the participants (case 5) was diagnosed with 16 pathological pockets (out of 56 examined).Table 3 presents the salivary characteristics. The mean stimulated secretion rate was 1.3 ml/min (SD 0.6, median 1.1), while the mean unstimulated secretion rate was 0.08 ml/min (SD 0.1, median 0.06). Four participants did not produce any unstimulated saliva. The majority (n = 12) had a medium or high salivary buffer capacity. The total bacterial count (206–3206) and the total number of streptococci (2.96–666) in the saliva were considerable. Four participants had a count of SM in the span 100,000–1,000,000 CFU/ml and six participants had ≥1,000,000 CFU/ml. Lactobacilli were present in the saliva from all participants, with half of them having a count of ≥100,000. As can be seen in Table 3, the bacterial count for most of the participants was high throughout; however, for two participants, no SM could be cultivated. Electrolyte concentrations were all within normal ranges; however, two participants scored high on fluoride content probably due to having brushed their teeth with fluoride toothpaste too close to the saliva collection.

TABLE 2. Dental caries, periodontal status and missing teeth Case Surfaces with caries (n) Filled surfaces Periodontal status Missing teeth Early stage decay (ICDAS 1–2) Established decay (ICDAS 3–4) Severe decay (ICDAS 5–6) % Plaque index % BOPa % Pocketsb n n 1 1 5 4 90.8 4.1 16.7 0 4 2 12 5 1 3.3 4.1 20.8 1 3 3 24 1 0 2.1 50.0 25.0 0 0 4 24 17 7 9.2 58.3 41.1 2 0 5 12 3 0 32.1 58.3 58.9 16 0 6 1 7 2 32.2 75.0 52.8 4 10 7 17 12 4 28.4 83.3 10.0 1 8 8 6 3 3 17.7 70.8 6.3 0 12 9 4 7 3 40.9 83.3 13.0 2 5 10 2 1 4 35.0 50.0 4.0 2 3 11 8 12 0 20.0 25.0 1.8 1 0 12 15 8 9 42.5 87.5 98.1 0 2 13 9 12 9 62.9 91.6 31.0 1 7 14 44 26 10 35.6 87.5 55.6 2 1 a Bleeding on probing. b Pathological periodontal pockets (≥3.5 mm). TABLE 3. Salivary characteristics: Production, buffering, microbiota and electrolytes Flow rate Microflora Electrolytes Case Unstimulated(ml/min) Stimulated (ml/min) Buffer capacity Bacteria (CFU/ml) Streptococci (CFU/ml) S. mutans (CFU/ml) Lactobacilli (CFU/ml) Calcium (mmol/L) Phosphate (mmol/L) Urea (mmol/L) Potassium (mmol/L) Fluor (μmol/L) 1 0 0.72 Low 1.3 × 108 4.4 × 107 1.9 × 106 9.2 × 106 1.12 3.80 4.10 22.40 26.05 2 0.13 1.32 Medium 1.1 × 108 3.8 × 107 2.0 × 103 1.4 × 104 1.15 5.20 5.70 26.90 2.48 3 0.04 0.84 High 2.0 × 107 2.9 × 106 1.9 × 104 1.3 × 104 0.84 4.00 4.00 17.40 25.16 4 0 2.16 High 7.4 × 107 1.4 × 107 0 1.8 × 105 0.91 4.20 3.10 23.20 2.47 5 0 1.50 High 9.2 × 107 2.0 × 107 1.6 × 106 1.5 × 104 0.09 4.00 2.70 25.40 1.88 6 0.09 2.56 Medium 1.2 × 108 2.7 × 107 2.6 × 105 1.2 × 104 0.97 3.50 3.80 19.90 1.72 7 0.08 1.32 High 1.8 × 108 5.8 × 107 2.3 × 106 3.0 × 104 0.92 3.10 3.90 18.20 1.56 8 0.15 1.00 Low 5.0 × 107 2.5 × 107 4.4 × 105 9.2 × 104 1.29 6.60 4.80 18.70 2.51 9 0 0.42 Medium 6.0 × 107 3.6 × 107 3.6 × 106 6.0 × 105 1.43 6.60 5.40 20.50 0.95 10 0.12 0.76 Medium 3.2 × 108 4.8 × 107 7.4 × 105 6.4 × 103 0.97 6.70 9.80 26.90 1.23 11 0.40 2.20 Medium 2.2 × 108 6.6 × 107 0 5.0 × 105 0.87 3.90 4.30 19.60 0.85 12 0.09 0.97 Medium 9.0 × 107 3.6 × 107 2.0 × 106 4.2 × 105 0.91 3.70 9.60 22.40 0.73 13 0.04 1.00 Medium 5.8 × 107 3.2 × 107 3.2 × 105 1.7 × 106 1.00 5.00 4.10 21.90 1.85 14 0.02 1.28 High 5.8 × 107 2.6 × 107 1.6 × 106 4.4 × 105 1.14 4.10 3.90 21.30 1.06

The outcomes in patient-reported variables are displayed in Table 4. The mean total energy intake was 1654 kcal/day (range 823–3074 kcal/day) with a mean of five meals per day. Four participants had a higher summed intake of sucrose than the Nordic nutrition council recommendation (2014), which is maximum 10% sucrose of the total energy intake.

TABLE 4. Patient reported outcomes, that is, dietary intake, behavior and self-perceived oral health Case Diet Oral health habits Oral health problems Oral health-related quality of life Mean intake of kcala Meal frequencya (n) Mean disaccharide intake (g)a Mean sucrose intake (g)