Bacterial overgrowth syndrome (BOS) is a term that describes clinical manifestations that occur when the normally low number of bacteria that inhabit the stomach, duodenum, jejunum, and proximal ileum significantly increases or becomes overtaken by other pathogens.[1]
The upper intestinal tract once was thought to be a sterile environment; however, low concentrations of various bacteria are now widely accepted to live within or attached to its luminal surface. These bacteria are thought to be present from the time of birth and through adulthood, living in symbiosis with the human host. This relationship is thought to be vital for normal digestive processes, immunity, and intestinal development. Bacterial species usually present include lactobacilli, enterococci, oral streptococci, and other gram-positive aerobic or facultative anaerobes.[2, 3, 4, 5]
Various etiologic processes can disrupt mechanisms that keep the number of these bacteria low. These include structural abnormalities (congenital or surgical) and disorders that cause decreased gastric acidity, reduced peristaltic activity, and mucosal damage or atrophy. The clinical manifestations of bacterial overgrowth syndrome stem from the increased bacterial burden on the normal functions of the upper GI system. Prompt recognition and treatment of bacterial overgrowth syndrome should be targeted to prevent and reverse malabsorptive processes.[6]
Structural changes in the stomach or small intestine can contribute to the stagnation of intestinal contents, leading to bacterial overgrowth. Conditions that necessitate or result in anatomical modifications, such as small-bowel diverticulosis, surgically created blind loops, postgastrectomy complications (particularly in the afferent loop of a Billroth II procedure), strictures, or partial blockages, can promote this bacterial overgrowth.
Moreover, motility disorders in the intestines linked to diabetic neuropathy, systemic sclerosis, amyloidosis, hypothyroidism, and idiopathic intestinal pseudo-obstruction can hinder the clearance of bacteria. In older individuals, factors like achlorhydria and spontaneous alterations in intestinal motility may also contribute to bacterial overgrowth.[1]
Small intestinal bacterial overgrowth (SIBO) commonly involves species such as streptococci, Bacteroides, Escherichia, Lactobacillus, Klebsiella, and Aeromonas.[6] The excessive bacteria in SIBO consume nutrients, leading to caloric deficiency and vitamin B12 deficiency, as they also deconjugate bile salts causing fat malabsorption and intestinal mucosa damage, resulting in diarrhea.[1]
Protective factors help maintain the upper gastrointestinal (GI) tract's bacterial balance, with abnormalities predisposing to bacterial overgrowth. Integrated motor processes like the migrating motor complex and migrating action potential complex clear bacteria and undigested substances from the upper intestine. Disruption of these processes can lead to diarrhea and weight loss, as anatomic defects can hinder peristaltic action.
Gastric acid and bile play roles in destroying microorganisms, whereas secretions from the intestine, pancreas, and bile aid in bacterial elimination in the small intestine. Mucosal integrity, mucin layer, immunoglobulins, and immune cells help protect the gut from bacterial invasion. Normal flora like Lactobacillus maintains a low pH to prevent overgrowth but can contribute to an abnormal luminal environment that allows the passage of enteric bacteria between the proximal and distal bowel.
Compromised ileocecal valve integrity can lead to increased microbial burden in the terminal ileum, resembling that in the colon, and impacting the absorption of bile acids, fats, carbohydrates, proteins, and vitamins. Malabsorption causes damage to the mucosal lining and enterocyte function, contributing to diarrhea and weight loss. Anaerobes like Bacteroides fragilis deconjugate bile acids, impairing their function, which leads to reduced fatty acid absorption and damage to intestinal cells. Unabsorbed sugars ferment, lowering pH and causing osmotic diarrhea, while deconjugated bile acids harm enterocytes and stimulate water secretion in the colon.[7]
The exact prevalence of bacterial overgrowth syndrome is likely underestimated because the clinical manifestations overlap with those of many other malabsorptive and diarrheal disorders. Higher clinical suspicion should be given to individuals with underlying disorders that disrupt the known protective elements that prevent bacterial overgrowth syndrome. For example, approximately 20-43% of chronic diarrhea cases in patients with diabetes, as well as 50% in neonates, may be associated with bacterial overgrowth syndrome.[8] In many cases, gastric and upper intestinal tract surgery results in bacterial overgrowth syndrome; however, preservation of the normal anatomy and antroduodenal vagal innervation appear to be protective.
Shah et al conducted a systematic review and meta-analysis on the prevalence of SIBO in patients with intestinal failure (IF) and identified potential risk factors for SIBO development. They reported a 57.5% prevalence of SIBO in IF patients, notably higher in those receiving parenteral nutrition (PN) and using PPI/acid-suppressing agents, but the evidence quality was limited by lack of case-control studies and clinical heterogeneity among included studies.[9]
However, Gandhi et al conducted a systematic review and meta-analysis focusing on the prevalence of methane-positive small intestinal bacterial overgrowth (SIBO) in patients with irritable bowel syndrome (IBS) and inflammatory bowel disease (IBD) compared to controls. This analysis revealed a methane-positive SIBO prevalence of 25.0% in IBS and 5.6% in IBD patients. While methane-positive SIBO was not significantly higher than in controls for IBS, it was associated with IBS-C and showed an inverse relationship with IBD, with Crohn's disease having lower prevalence than ulcerative colitis. The study emphasized the need for further research due to low evidence quality resulting from clinical heterogeneity.[10]
Moreover, the prevalence of bacterial overgrowth syndrome (BOS) can vary depending on the population used and the diagnostic methods applied. In healthy individuals, BOS rates range between 0-12.5% by the glucose breath test, 20-22% by the lactulose breath test, and 0-35% with the C D-xylose breath test.[11] BOS tends to be more common in the elderly population due to factors such as decreased gastric acid secretion and consumption of multiple medications that can lead to hypomotility, with BOS being linked to occult malabsorption in this demographic.[12]
Bacterial overgrowth syndrome can lead to worsening symptoms of malabsorption and diarrhea. In certain patient subgroups, bacterial overgrowth syndrome can lead to significant morbidity or death. However, exact mortality rates directly linked to bacterial overgrowth syndrome are not readily available.
Patient populations at an increased risk for mortality due to bacterial overgrowth syndrome include the following:
If bacterial overgrowth syndrome is the result of an underlying medical problem that cannot be controlled, relapse will occur, with symptom-free periods.
Patients with chronic diarrhea should be educated on avoidance of food products that may exacerbate symptoms. Patients with bacterial overgrowth syndrome should document which foods cause their diarrhea, as this can vary among patients. Some examples of such foods are those high in carbohydrates such as fruits and fruit juices, spicy food, milk-containing products, fried food, and high-fat foods.
Patients also should be educated on early detection of symptoms such as diarrhea to avoid malabsorption.
In high-risk patients (eg, neonates and elderly patients), early recognition is challenging. Education should be extended to the primary care givers in this situation.
No specific symptoms are pathognomonic for bacterial overgrowth syndrome (BOS). Nonetheless, various nonspecific GI symptoms are common in affected individuals. Clinicians should have a heightened clinical suspicion for bacterial overgrowth syndrome in patients who present with the following[1] :
Advanced cases of bacterial overgrowth syndrome may manifest as malabsorption findings, as follows:
Disorders or structural abnormalities that disrupt the protective mechanisms that guard against increasing bacterial burden can lead to bacterial overgrowth syndrome.
Patients with the following medical conditions are at increased risk for bacterial overgrowth syndrome:
Abnormal small intestinal motility due to the following may result in bacterial overgrowth syndrome:
Blind pouches from the following may result in bacterial overgrowth syndrome:
Abnormal bowel communication due to the following may cause bacterial overgrowth syndrome:
Partial obstruction caused by the following may result in bacterial overgrowth syndrome:
Reduced gastric acid secretion from the following may result in bacterial overgrowth syndrome:
Prevalence of BOS rises with age.[11, 18]
A complete physical examination should be performed with emphasis on abdominal examination and examination for signs of malabsorption of various nutrients. No specific physical examination techniques are required for bacterial overgrowth syndrome.
Complications of bacterial overgrowth syndrome are possible in patients with prolonged and untreated symptoms, potentially leading to increased morbidity and mortality in higher-risk patients (eg, the very young and elderly).
Malabsorption of iron, vitamin B-12, and folate can lead to anemia
Persistent diarrhea can lead to volume loss and electrolyte disturbances.
Decreased fat absorption can lead to further diarrhea.
Bacterial overgrowth syndrome (BOS) diagnostic testing should include a workup for diarrhea, anemia, and malabsorption.[1] In the past, retrieval of aspirates from the small intestine itself during endoscopy was the diagnostic tool of choice; however, its use was limited due to low specificity.
Standard anemia workup and nutritional evaluation are indicated.
Stool analysis can help detect abnormal stool components.[19] The pH may be acidic, and reducing substance may be present in the stool.
D-lactic acidosis syndrome can result from carbohydrate fermentation. Lactic acid levels may need to be measured and, if elevated, monitored. D-lactic acid levels, measured in the blood or urine, can help differentiate bacterial overgrowth syndrome from other metabolic causes.[20]
Short-chain fatty acid levels may be elevated in the duodenal fluid but not the stool.[21] Abnormal duodenal short-chain fatty acid levels average approximately 1 µmol/mL, with acetic acid, propionic acid, and n -butyric acid representing 61%, 16%, and 8% of the total, respectively. The average short-chain fatty acid level in a healthy patient is 0.27 µmol/mL, with acetic acid representing 84% of the total.
Keto-bile acid concentration in duodenal fluid is increased and correlates with the type of bacterial overgrowth.[22] The molar percent of keto-bile acids in normal duodenal fluid is very close to 0, while gram-negative aerobic and anaerobic overgrowth is associated with levels of 32 µmol/mL and 11 µmol/mL, respectively.
Urine 4-hydroxyphenylacetic acid levels may be elevated.[23] Enteric bacteria that possess L-amino acid decarboxylase produce 4-hydroxyphenylacetic acid from dietary tyrosine. Increased excretion has been demonstrated in adults with bacterial overgrowth syndrome. Creatinine levels that exceed 120 mg/g are typical in children with small-bowel disease or bacterial overgrowth syndrome, including children with chronic Giardia lamblia gastroenteritis. Children with severe pancreatic dysfunction secondary to cystic fibrosis also have significantly high levels of this metabolite. A 2% false-positive rate and no false-negative results are found when this test is used to screen healthy control subjects and hospitalized children.
In cases where anatomic alterations are not a result of previous surgery, the identification of predisposing anatomic lesions can be achieved through an upper gastrointestinal series with small-bowel follow-through, CT enterography, or magnetic resonance enterography. These tests can help detect structural abnormalities such as strictures, malrotation, diverticulosis, fistulae, and pseudo-obstruction. If upper gastrointestinal evaluation fails to provide a diagnosis, imaging and examination of the lower GI tract should be considered.[1]
Breath testing, particularly using glucose hydrogen or lactulose hydrogen, is recommended for diagnosing small intestinal bacterial overgrowth (SIBO) in symptomatic patients. Before undergoing breath testing, patients are advised to refrain from antibiotics for at least 4 weeks and avoid promotility agents and laxatives for at least 1 week.[1]
A positive breath test for SIBO is identified by a greater than 20-ppm increase in hydrogen within the first 90 minutes or a more than 10-ppm increase in methane at any point during the test.[1] These tests are crucial for detecting the byproducts of bacterial metabolism and identifying malabsorbed substances in the gut.[24] While multiple studies support the specificity of three breath tests, the sensitivity remains a topic of varying agreement among researchers. Research comparing these tests with duodenal bacterial counts indicates that the xylose breath test offers the highest level of specificity.[25]
Takakura et al explored the reliability of a fasting single methane measurement (SMM) as a diagnostic tool for intestinal methanogen overgrowth (IMO) and its role in monitoring treatment response. They compared SMM with traditional breath tests, analyzed associated symptoms, and assessed the stability of SMM in a randomized control trial. The study revealed that an SMM level of ≥10 ppm showed high sensitivity (86.4%) and specificity (100%) for IMO diagnosis, with a noteworthy link to constipation. Stability of SMM levels without treatment and decreased post-antibiotic therapy, along with a positive correlation with stool M. smithii load, suggest SMM's efficacy in diagnosing and monitoring IMO, making it a valuable diagnostic and treatment evaluation tool.[26]
Hydrogen breath tests are based on the fact that in humans hydrogen is exclusively produced by intestinal bacteria, most notably by anaerobic bacteria in the colon of healthy people and also in the small intestine in the case of bacterial overgrowth syndrome.[27] Preoral glucose or lactulose challenge is given before performing hydrogen breath tests. Bacteria ferment malabsorbed carbohydrates. Fermentation releases hydrogen gas that is absorbed and excreted by the lungs.
Under normal conditions, fermenting bacteria reside in the colon. In bacterial overgrowth syndrome, the exhaled hydrogen concentration rises early, corresponding to small intestinal bacteria fermentation of carbohydrates. Under such conditions, a later rise in exhaled hydrogen may also be detected during large bowel fermentation. Antibiotic administration invalidates this test.
For diagnosis, use 1-2 g/kg of glucose, not to exceed 25-50 g. A rise in exhaled hydrogen to 20 parts per million is diagnostic. For diagnosis, use 10 g lactulose. A rise in 20 parts per million above baseline is diagnostic. The specificity and sensitivity of this test are 62.5% and 82% after glucose and 56% and 86% after lactulose administration.[28]
Give glycocholate tagged with carbon 14 with a light meal, and collect breath samples at 2, 4, and 6 hours. An abnormal rise in radioactive carbon dioxide levels indicates bacterial deconjugation of glycocholate.
The specificity and sensitivity of this test are 60-76% and 33-70%, respectively
False positive results may come from disease or resection of terminal ileum, the site of bile absorption. Carbon 14 carries a risk of radiation, which can be problematic in children and pregnant women.[11]
Gram-negative bacteria metabolize xylose, resulting in the release of radioactive carbon dioxide.[27] Administer 1 g of D-xylose tagged with carbon 14, as a liquid, after an overnight fast. Measure radioactive breath carbon dioxide at 30, 60, 90, and 120 minutes. An abnormally high carbon dioxide concentration usually is detected within 30-60 minutes. The specificity and sensitivity of this test are 14.3-95% and 40-94%, respectively.[11]
Combination of hydrogen breath test with simultaneous D-xylose breath test results in increase in sensitivity of noninvasive diagnostics of bacterial overgrowth syndrome.[29, 30]
Descending duodenal biopsies performed in a group of elderly individuals with bacterial overgrowth syndrome demonstrated that mean villus height, mean crypt depth, and total mucosal thickness may be reduced.[31] These indices are not significantly different from controls after 6 months of treatment of bacterial overgrowth syndrome. A significant drop in the number of intraepithelial lymphocytes is also seen over this observation period. Mucosal atrophy can result in an 80% reduction of intestinal surface area in infants. Once the offending agent is removed, repair of the small bowel progresses slowly. After 2 months, the villi surface area is 63% normal but the microvillous surface area is only 38% normal.
The standard diagnostic method for small intestinal bacterial overgrowth (SIBO) involves quantitatively culturing intestinal fluid aspirate to identify a bacterial count greater than 10^3 colony-forming units/mL.[1] This approach typically requires endoscopy for sample collection. While quantitatively measuring bacteria aspirate from sterile cultures obtained from the proximal small bowel can aid in Bacterial Overgrowth Syndrome (BOS) diagnosis, a bacterial colony count equal to or exceeding 10^3 colony-forming units per milliliter (CFU/mL) is now considered diagnostic according to recent North American Consensus guidelines,[32] contrary to the previous threshold of 10^5 CFU/mL.[33] Despite its effectiveness, this technique is costly and invasive, prompting consideration of simpler diagnostic methods as a first step.
In the management of bacterial overgrowth syndrome (BOS), antibiotic treatment may be cyclic and adjusted based on culture and sensitivity results if symptoms recur, although changing antibiotics can be challenging due to the presence of multiple bacterial species. Dietary modifications like a high-fat, low-carbohydrate, and low-fiber diet can be beneficial as bacteria primarily metabolize carbohydrates in the intestinal lumen.[34, 35, 36, 37, 38]
Additionally, correcting underlying conditions and nutritional deficiencies, such as vitamin B12 deficiency, is essential. Treatment of BOS involves addressing the primary underlying disease, if present, with antibiotic therapy and nutritional support. Management aims to treat any disease or anatomic defect that contributed to bacterial overgrowth. While many associated clinical conditions may be irreversible, antibiotic therapy is utilized to rebalance enteric flora, targeting symptom reduction without completely eradicating protective microorganisms. Initial antibiotic therapy is typically broad-spectrum to cover both aerobic and anaerobic organisms, considering community resistance patterns.[1]
Rifaximin is currently the mainstay of treatment. Tetracycline was the mainstay of therapy, but its use as a single agent has fallen out of favor in adult patients given community increases in bacterial resistance.
Bacterial sensitivities from duodenal intubations with nonidiopathic bacterial overgrowth syndrome support the use of amoxicillin-clavulanate. Amoxicillin-clavulanate appears to be 75% effective in patients with diabetes.
Takakura et al conducted a systematic review and meta-analysis on the effectiveness of antibiotics in treating small intestinal bacterial overgrowth (SIBO) and irritable bowel syndrome (IBS) patients with and without SIBO. The analysis found that antibiotics were significantly more effective in improving symptoms in patients with SIBO compared to those who received placebo or no antibiotics. Additionally, IBS patients with SIBO showed higher response rates to antibiotics than those without SIBO. These results suggest a potential benefit of antibiotics in treating SIBO and highlight the importance of considering precision medicine approaches in IBS treatment. However, larger, multicenter studies are needed to validate these findings, given the limitations of small sample sizes and data quality in the current studies.[39]
Studies show that rifaximin eradicates bowel overgrowth syndrome in as many as 80% of patients.[40, 41] Higher doses (1200 or 1600 mg/d) are more effective than standard doses (600 or 800 mg/d).[42] Long-term favorable clinical results have been achieved with rifaximin in patients with irritable bowel and bacterial overgrowth syndrome.[43]
Clindamycin and metronidazole are useful in elderly patients with idiopathic bacterial overgrowth syndrome.
Cholestyramine reduces diarrhea in infants and neonates with intractable diarrhea.[44] Infants with 10-25 days of severe persistent diarrhea for which a cause could not be found despite an extensive infectious and immunologic workup were treated with cholestyramine and gentamicin or metronidazole. Cholestyramine and gentamicin significantly reduced stool weight within 4-5 days of therapy but had mild detrimental effects on fat and nitrogen absorption.[45]
Ciprofloxacin and metronidazole result in normalization of hydrogen breath tests in most patients with Crohn disease.[14]
Norfloxacin, cephalexin, trimethoprim-sulfamethoxazole, and levofloxacin have been recommended for the treatment of bacterial overgrowth syndrome.[11, 46, 47]
The exact length of therapy is not clearly defined; length of therapy should be tailored to symptom improvement. A single 7-10 day course of antibiotic may improve symptoms in 46-90% of patients with bacterial overgrowth syndrome.[48] Recurrence following therapy is not uncommon and is more likely in elder patients, especially those with history of appendectomy and chronic proton pump inhibitor use. Patients with recurrent symptoms may need repeated (eg, the first 5-10 d of every month) or continuous use of cyclical antibiotic therapy.[11]
Alternative Therapies
Nickles et al conducted a systematic review on the use of alternative therapies, such as probiotics, therapeutic diets, and herbal medicines, in the treatment of SIBO. Although some studies suggest a potential benefit, the overall evidence is limited and lacks robust clinical trials. Existing studies vary in design and lack standardization, indicating the need for large-scale, placebo-controlled trials to determine the effectiveness of alternative therapies for SIBO treatment.[49]
Probiotic therapy results in bacterial overgrowth syndrome have been inconclusive and generally not recommended for general clinic use.[7, 50]
Therapeutic use of prokinetics in bacterial overgrowth syndrome due to motility disorders have been tried in many studies. Metoclopramide, cisapride, domperidone, erythromycin, tegaserod, and octreotide have been used; however, data suggest long-term effectiveness is limited.[46]
Nutritional support with dietary modifications such as lactose-free diet, vitamin replacement, and correction of deficiencies in nutrients like calcium and magnesium should be an important part of bacterial overgrowth syndrome treatment, if applicable.
Certain potential underlying abnormalities are amenable to treatment, as follows:
The following potential underlying diseases are not amenable to treatment, but prevention of their progression may be therapeutic:
In the absence of underlying structural abnormalities that limit normal bowel function, surgery generally is unwarranted.
Repair postoperative strictures and blind loops; for example, a Billroth type II may need conversion to a Billroth type I.
Strictures, fistulae, and diverticula may require surgical correction.
Patients refractory to standard medical or surgical treatment or those who have severe symptoms should be referred to a gastroenterologist/infectious disease specialist for additional workup.
No robust studies have yet properly defined the role of dietary modification in the treatment of bacterial overgrowth syndrome. A low-fiber diet and reduction of fermentable sugars such as sucralose may reduce the occurrence of bacterial overgrowth syndrome, although this potential is extrapolated from results in patients with IBS. There is growing interest in the role of low-FODMAP (fermentable oligo-, di-, mono-saccharides and polyols) diets in bacterial overgrowth syndrome. Studies on the role of FODMAP in patients with IBS found that a low-FODMAP diet was associated with fewer fermentation products, as assessed with a breath test.[51, 52]
Controlling the underlying etiology is an important component of preventing bacterial overgrowth syndrome recurrence. Known underlying medical conditions that predispose to bacterial overgrowth syndrome (eg, diabetes mellitus, scleroderma, alcoholism, cirrhosis, chronic pancreatitis, hypochlorhydria due to atrophic gastritis or medications) need to be optimally controlled. Surgical intervention may be beneficial in patients with anatomic abnormalities such as surgically created blind loops, strictures due to previous surgeries, and conditions such as Crohn disease who continue to have persistent symptoms of bacterial overgrowth syndrome, including unintentional weight loss, diarrhea, and bloating, among others, after adequate treatment. Patients with motility disorders should avoid medications that decrease intestinal motility.
Given the high relapse rate associated with bacterial overgrowth syndrome following antibiotic therapy, some clinicians opt to repeat a second course of antibiotics to reduce the risk for early relapse. A small study suggested that patients who undergo retreatment with antibiotics tend to have a lower relapse rate, although no clear guidelines support this.[53]
Antibiotic prophylaxis usually is reserved for patients who experience multiple relapses (≥4) per year and who are known to have risk factors for recurrence.
The use of probiotics to treat or prevent bacterial overgrowth syndrome has not been well validated. Limited studies support their use. A meta-analysis of small studies found no difference in the incidence of bacterial overgrowth syndrome among the probiotic group versus the control group.[54]
Close-interval follow-up is recommended to ensure that therapy is improving symptoms. It is also not clearly defined whether serial testing for increased bacteria burden is warranted.
Admission criteria for bacterial overgrowth syndrome (BOS) should be based on severity of clinical manifestations at presentation, especially in high-risk individuals.
Only recently have well-established guidelines for the management of bacterial overgrowth syndrome (BOS) been made available. Treatment has been mostly empirical and extrapolated from diseases with similar clinical presentations, including irritable bowel syndrome, among others.
In 2017, “Hydrogen and Methane-Based Breath Testing in Gastrointestinal Disorders: The North American Consensus” was published and focuses mainly on the diagnosis of suspected bacterial overgrowth syndrome using hydrogen and methane breath tests.[27]
No clinical trials have been conducted to study more effective treatment methods for bacterial overgrowth syndrome. They are summarized below:
In February 2020, the American College of Gastroenterology established comprehensive guidelines for the diagnosis and management of small-intestinal bacterial overgrowth. They are summarized below[32] :
The goals of pharmacotherapy are to reduce the burden of microorganisms in the intestine rather than achieve complete eradication, to reduce morbidity, and to prevent complications.[1]
Clinical Context: Antibiotic of choice as clinical resistance may be less frequent than with other antibiotics.
Clinical Context: First-line antibiotic for bacterial overgrowth syndrome due to anatomic abnormalities and diabetes and for elderly patients with idiopathic bacterial overgrowth syndrome. Provides good gram-negative, gram-positive, and anaerobic coverage. Reduces number of bacteria in small bowel lumen.
Clinical Context: Works well in elderly patients with idiopathic bacterial overgrowth syndrome, especially if bile malabsorption coexists. Good anaerobic and gram-positive coverage, except enterococci.
Clinical Context: First-line antibiotic for elderly patients with idiopathic bacterial overgrowth syndrome. Provides good anaerobic coverage.
Clinical Context: Effective for patients with bacterial overgrowth syndrome. Provides anaerobic coverage.
Clinical Context: Trimethoprim-sulfamethoxazole can be used as alternative antibiotics in both pediatric and adult patients with bacterial overgrowth syndrome. The dose should be adjusted in patients with renal impairment. Sulfamethoxazole and trimethoprim cross the placenta and should be avoided in pregnant women to prevent increased risk of congenital malformations.