Complete-Case Sedation Dentistry

A mouth treated one emergency at a time

An adult in middle age arrives with a folder of earlier treatment plans and a long history of not being able to sit in a dental chair. The fear goes back to childhood. The medical history has accumulated since: type 2 diabetes, a blood pressure that is controlled on two agents, a body mass that makes the airway something to think about, and snoring that a partner has complained about for years. Over the past decade the patient has been sedated twice for dental care, each time to deal with whatever hurt most that month. Each of those appointments went smoothly. Each dealt with the tooth it was booked for. And in between, the rest of the mouth went on doing what untreated disease does.

Every one of those earlier plans was a reasonable plan for the problem in front of it. A painful tooth is a real emergency, and a sedation appointment that relieves it has done its job. The difficulty is not with any appointment. It is that the sequence was never designed as a sequence. Nobody had set out, at the start, to finish this mouth — and a patient who can only tolerate care under sedation, and only rarely, cannot afford a workflow that treats the mouth one problem per event.

This piece sets out the approach I take with patients like this one, and why. It is not complicated to describe. Treat the whole mouth definitively, in one planned event, scoped from the outset to finish it. Plan any case that will be complete, or that will need deeper sedation, for the hospital or for an anaesthesiologist from the beginning. Keep office cases limited in scope from the start, and do not expand them partway through. The reasons for each of those come from the same mechanism, and the mechanism is what this piece is mostly about.

Who this is for

The patients I have in mind are anxious and medically complex patients who need complete care. That is a broad heading on purpose. It includes the person with lifelong dental fear who has avoided care until the disease is extensive; the patient with cardiac, respiratory or metabolic disease whose reserve narrows what can be done in a chair; the patient who cannot cooperate with care because of a developmental or cognitive condition; and the patient whose previous attempts at treatment have failed and who now associates the dental setting with that failure. Special-needs dentistry is one lane under that heading, not the label for it.

What these patients share is not a diagnosis. It is that every treatment event costs them something that a routine patient does not pay — fasting, an escort, a day lost, an anaesthetic or sedation risk that is higher than average, and the anticipation that comes before each appointment. When each event is expensive, the number of events becomes a clinical variable in its own right.

The cost of each event

A sedation or anaesthetic event carries a fixed cost that does not shrink because less dentistry is done during it. The fasting is the same. The escort is the same. The pre-sedation assessment, the monitoring, the recovery and the hours afterwards are the same. The airway risk scales more with the patient and the depth than with the number of teeth treated. So when a mouth that needs twenty procedures is treated four procedures at a time, the patient pays the fixed cost five times — and that assumes nothing new develops in the intervals, which in an active-caries mouth it usually does.

There is a second cost that does not show up in a risk table. For an anxious patient, a failed or difficult attempt makes the next one harder. A bad experience in one setting colours how the patient approaches the next, and the general anaesthetic that follows failed attempts is a much more difficult event than the one that might have been planned first. That is mitigable — agents with anterograde amnesic properties, a change of surroundings so the next attempt is not made in the room the patient has learned to fear — but a first failure is recoverable rather than free.

The aggressiveness of treatment in one event follows from that. I weigh it roughly inversely to how easily the patient could be brought back. A combative, large, irritable patient with extensive caries gets more definitive treatment, specifically so that another anaesthetic is not needed soon. A patient who goes to sleep easily and carries less risk is one for whom conservative measures — sealants, fluoride, watching a lesion — are reasonable to attempt, because the next visit is cheap. The factors being balanced are the caries risk, the anaesthetic risk, the patient's own medical risk, and what the patient actually needs.

What the repeat-anaesthetic literature says

Most of the outcome evidence for treating the whole mouth in one anaesthetic comes from children with early childhood caries, because that is where dental general anaesthesia is most studied. The adult and special-needs evidence is thinner, and that limitation applies to almost every number in this section.

Two things come through. The first is that a substantial share of children need a second dental general anaesthetic. A ten-year review at a children's hospital in Toronto found a repeat rate of 10.8%, and that the rate rose over the decade; having a medical comorbidity was not associated with repeat treatment (Vertullo and colleagues, 2021). A six-year follow-up of 484 children who had extractions under general anaesthesia at a UK teaching hospital found a repeat rate of 10.7%, with caries the reason in 84% of repeats — and 72% of the teeth extracted in the years afterwards had been recorded as caries-free or unerupted at the first anaesthetic (Kakaounaki and colleagues, 2006). That last figure matters. Much of the later treatment in that series was for disease that had not been present at the first event, rather than for work left undone, which is a reminder that the event alone does not settle the outcome; prevention afterwards does as much.

The second is that caries recurrence after comprehensive treatment is common. In one hospital cohort, 53% of children with early childhood caries had new lesions within two years of comprehensive treatment under general anaesthesia, and those who missed their immediate follow-up were more likely to relapse (Foster and colleagues, 2006). Only 193 of the 448 children scheduled for treatment met the study's follow-up criteria, so this is a single-centre retrospective figure with large attrition, and the true rate could be higher or lower.

What lowers the repeat rate is more interesting than the rate itself. A matched case-control study of young children found that full-coverage restorations on primary molars and canines at the first anaesthetic were associated with lower odds of a second, while resin restorations and sealants on second primary molars were associated with higher odds (Azadani and colleagues, 2024). And a children's hospital service evaluation reported a repeat rate within two years of 0.63%, which its authors attributed to specialist treatment planning and to running separate lists for extraction-only and restorative cases (Kirby and colleagues, 2020). That is the authors' explanation of their own figure rather than a tested comparison, and repeat rates are not comparable between services that define and count them differently — but it is the clearest published statement of the idea this piece rests on, which is that the repeat rate is partly a planning outcome.

A review of dental treatment under general anaesthesia for patients with special needs located only ten studies, of which two reported repeat rates — 7.2% and 10.2% — and noted that there were no accepted protocols for providing this care (Mallineni and Yiu, 2016).

Complete or staged: what has and has not been compared

The obvious question is whether finishing the mouth in one event produces better outcomes than staging it across several sedation visits. It has not been answered by a trial. The Cochrane review comparing sedation with general anaesthesia for dental treatment in patients under 18, with searches run to July 2015, found no eligible randomised trials at all (Ashley and colleagues, 2015). A systematic review of pharmacological versus non-pharmacological approaches for adults with special needs, searched in April 2024, found only two eligible studies and could not reach a conclusion (Vacharaksa and colleagues, 2025). A PubMed search I ran in November 2026 did not locate a randomised comparison of single-event comprehensive treatment against staged treatment under sedation in any population.

What exists is indirect. When a tertiary paediatric dental clinic added an in-office general anaesthesia service alongside its oral sedation programme, case completion rose from 88.6% under oral sedation to 99.5% under general anaesthesia, complete radiographs were obtained in every anaesthetic case against 63.4% of oral sedation cases, and one serious adverse event occurred in 418 anaesthetics against none in 508 oral sedations (Moore and colleagues, 2019). That is a before-and-after comparison in one institution, with different patients selected for each route, and it is not evidence that one route is safer. What it does show is the practical point behind scoping: a plan to finish the mouth depends on being able to see the whole mouth first, and on an event long and deep enough to do the work.

The adult evidence is sparser still. A five-year review of 363 adults with special needs who had comprehensive dental rehabilitation under general anaesthesia in an outpatient operating room recorded two complications, one airway-related and one surgical; both led to unplanned admission and both resolved without residual harm (Messieha and colleagues, 2007). That is a reassuring series and a small one.

Scope from the outset

Planning the whole case at the start is what makes everything else coherent. Scoping means the examination, the radiographs and the diagnosis are complete before the treatment event is booked, so the event can be sized to the work rather than the work to the event. It is also what decides the venue. A case that will be complete, or that is likely to need deeper sedation to be completed, goes to the hospital or to an anaesthesiologist from the outset — not arranged partway through when the office case turns out to be harder than expected.

I write that as following from the mechanism rather than as a rule for anyone else. The decision is a balance: the risks and problems of each route weighed against patient safety and against the sequelae of getting it wrong in either direction. A patient can be sent to the operating room who would have done well in the chair, and a patient can be kept in the chair who should have gone to the operating room. Both are errors, and both have costs.

Office cases are limited in scope from the start, and they do not expand mid-appointment. The reason is the sedation itself. A sedation plan is built for a depth and a duration. Extending the case means more drug, more time and a longer recovery than the plan assumed, in a patient whose response to the earlier increments has already told me something about where the margin is. What follows from that is that the second half of a mouth found partway through an office appointment belongs to a new plan, not to the appointment in progress.

How I sedate

The other half of the approach is how the sedation itself is given. What I protect is the ability to hold the patient at moderate sedation and to wake them. Everything else follows from that.

The American Society of Anesthesiologists' multisociety guideline for moderate sedation, written jointly with dental organisations including the American Dental Association and the American Society of Dentist Anesthesiologists, treats sedation as a continuum: minimal sedation, moderate sedation, deep sedation and general anaesthesia, distinguished by how the patient responds and whether the airway, breathing and circulation are maintained without help (American Society of Anesthesiologists Task Force, 2018). The paediatric guideline from the American Academy of Pediatrics and the American Academy of Pediatric Dentistry builds its safety requirements on the same idea — that the clinician must have the skills and equipment to rescue a patient from a level deeper than intended (Coté and colleagues, 2019). A patient does not stay at the level the plan names. Depth is a state the patient is in, not a property of what was given, and patients move between states.

So I give very little at a time. If a case goes deeper than intended, I stop giving medication and let the drug's offset bring the patient back. I avoid agents whose half-lives or metabolites accumulate, because an agent that accumulates makes the offset slower and less predictable, and a slow offset is exactly what I am relying on not to have.

I use midazolam and triazolam only. That is not a claim that other agents are unsafe; it is that I know these two drugs' characteristics and their limits, and in a technique that depends on predicting where the patient is going next, knowing the drug is the technique.

Why I do not use fentanyl

I will not use fentanyl. That is not a criticism of the drug, which is used widely and well. It is that with an opioid and a benzodiazepine together, when a patient's airway begins to fail I cannot tell whether what I am seeing is reduced respiratory drive or reduced muscle tone in the throat — and those have different answers. A patient who has stopped trying to breathe needs to be reminded or assisted to breathe. A patient who is trying to breathe against a soft airway needs the airway opened. With one class of drug on board, the picture is easier to read.

The interaction is well documented. In a volunteer study, midazolam alone produced no significant respiratory effects; fentanyl alone produced hypoxaemia in half the subjects but no apnoea; and the combination produced hypoxaemia in 11 of 12 subjects and apnoea in 6 of 12 (Bailey and colleagues, 1990). In a placebo-controlled trial in 207 young adults having third molars removed under intravenous sedation, 63% of those given midazolam with fentanyl became apnoeic against 3% of those given midazolam alone (Milgrom and colleagues, 1993). That same trial found the combination group more than four times as likely to have excellent sedation at a given level of pain, which is the honest reason the combination is popular, and a cost of declining it that belongs in the record.

A retrospective review of 61,237 open-airway intravenous sedation cases in oral and maxillofacial surgery offices found 25 adverse events, an incidence its authors reported as 3 per 100,000 per year, and a statistically significant association between adverse events and fentanyl (Wiemer and colleagues, 2024). With 25 events, adverse events identified through surrogate markers in the chart, and univariate tests that cannot separate the drug from the cases in which it was chosen, that is an association rather than a measured effect of the drug, and the study's own headline is how rare events were.

Capnography

Capnography is not required under my permit, and I do not hold that its absence leaves me blind. Chest rise, the patient answering, asking whether they are all right, and the other clinical signs give me much of the same feedback about ventilation. That is my position, and the evidence on it is genuinely mixed — but where it is strongest, it is strongest in exactly the kind of setting I work in, and that needs saying plainly.

The meta-analyses point one way. A meta-analysis of 13 randomised trials of procedural sedation found that adding capnography to visual assessment and pulse oximetry reduced mild and severe desaturation and the use of assisted ventilation, with no significant difference in the other endpoints studied (Saunders and colleagues, 2017). A rapid review of 16 studies published in the Journal of the American Dental Association, restricted to adults under moderate sedation, found that capnography detected adverse respiratory events with higher sensitivity and may reduce hypoxaemia by about 31% — on evidence it rated moderate to low quality — while not affecting serious adverse events, sedation quality or satisfaction (Parker and colleagues, 2018). The trials behind these pooled estimates come largely from medical procedural sedation rather than dentistry.

The one dental trial of capnography under conscious sedation with midazolam was negative. In 190 patients having oral surgery, randomised to whether staff could see the capnograph, hypoxaemia occurred in 34.4% of the capnography group and 39.2% of the control group, a difference that was not statistically significant (Brady and colleagues, 2017). Two details limit what it can say: patients were not routinely given supplemental oxygen, and at 190 patients it cannot speak to rare serious events. The only other dental randomised trial was positive — hypoxaemia in 13.4% against 34.8% — but it concerned deep propofol sedation given by dental anaesthesiologists to patients with intellectual disabilities, which is a different technique at a different depth (Takaya and colleagues, 2017).

The reason the published support is strongest in a setting like mine is mechanical. Pulse oximetry detects hypoventilation late when the patient is breathing supplemental oxygen; in a study of deliberate hypoventilation, saturation fell only in patients breathing room air (Fu and colleagues, 2004). Where a sedated patient is breathing supplemental oxygen, that lag matters most. The dentist working alone is watching the mouth rather than the chest, often with a drape, a rubber dam and the patient's own posture obscuring the signs. And the patients this piece is about are often the ones least able to say that something is wrong. Those are the conditions under which an early ventilation signal earns its place, and the 2018 multisociety guideline recommends monitoring ventilation with capnography during moderate sedation unless precluded by the patient, the procedure or the equipment (American Society of Anesthesiologists Task Force, 2018). My practice departs from that recommendation, and I would rather say so than leave it implied.

The day itself

For a hospital case, younger children and patients who cannot wait for any reason are scheduled first. Where the hospital's and the anaesthesia team's protocols allow, a parent or caregiver can be in the room for induction and is escorted out once the patient is asleep. The plan written at the scoping visit is the plan carried out.

Afterwards, the event is not the end of the plan. In the repeat-anaesthetic studies above, new disease after comprehensive treatment was common, and missing the first follow-up was associated with relapse. A complete case buys a patient a clean start, and what happens next decides whether it lasts. That is an argument for settling recall and preventive care with the patient or caregiver as part of the plan, before discharge rather than after it.

Where I could be wrong

The weakest part of this argument is its evidence base. Almost all of the repeat-anaesthetic and recurrence data come from young children with early childhood caries; I am applying them to adults with dental fear and medical complexity because that is where I work, and the inference is mine. Nobody has randomised comprehensive single-event treatment against staged treatment, and the indirect comparisons are confounded by who gets selected for each route.

Longer events have costs of their own. In a prospective study of children treated under general anaesthesia, 82% reported post-operative dental pain, and pain was associated with the number of teeth treated (Hu and colleagues, 2018). A single long event concentrates that. The trial evidence that general anaesthesia does not harm neurodevelopment comes from infants anaesthetised for a median of 54 minutes (McCann and colleagues, 2019), which says little about a long anaesthetic for full-mouth treatment in a young child. And in the only matched comparison located of dental general anaesthesia in autistic and neurotypical children, the autistic group had shorter anaesthetics but a considerably higher perioperative complication rate — 32.1% against 7.6% in one university hospital's records (Ünal and colleagues, 2026). A single-centre figure is not a general rate, but the direction matters: routing a patient to the operating room does not make the event straightforward, and the case for one planned event is a case for planning it carefully, not for assuming it is benign.

It is also not clear that general anaesthesia outperforms careful chairside care on durability. A comparison of over a thousand procedures in 92 children found no significant difference in the need for re-treatment between general anaesthesia and the chair, with restoration failure from secondary caries prominent enough that the authors flagged it (Bağ and colleagues, 2023). And a meta-analysis found no evidence that dental general anaesthesia either increases or decreases dental anxiety in children, on evidence rated very low certainty (Keykha and colleagues, 2025) — which cuts against any claim that one completed event resets a fearful patient.

On venue, one finding is often quoted and needs its denominator. A critical incident analysis of 95 paediatric adverse sedation events found death and permanent neurological injury more frequent in non-hospital settings, with inadequate resuscitation rated a determinant more often there (Coté and colleagues, 2000). It has no denominator of cases performed in each venue, so it does not measure the risk of a venue; it is a series of reported incidents, drawn from adverse-event reporting systems and a survey of specialists. It is a reason to plan the rescue, not evidence that one setting is safe and another is not.

On my own practice, the capnography position is where the published standard and I part company most clearly, and the setting where I work is the setting where the argument for it is best. Declining fentanyl gives up sedation quality that the dental trial evidence shows is real. And the rule about not expanding office cases will sometimes mean a patient comes back for a second event that, on the day, could have been avoided. I accept that cost because the alternative is deciding the depth of a sedation partway through it, but it is a cost, and a reader who weighs it differently is not wrong to. Where a reader's experience differs, I would like to hear it.

Sources

Evidence located via PubMed. Guidelines, systematic reviews and meta-analyses first; DOI links where a DOI exists, PubMed record otherwise.

1. Vertullo L, Barrett E, Quinonez C, et al. Trends in repeat general anaesthesia for treatment of dental caries at a children's hospital in Toronto, Canada: a 10-year retrospective investigation. Eur Arch Paediatr Dent. 2021;22(6):1087-1093. 10.1007/s40368-021-00667-6
2. Kakaounaki E, Tahmassebi JF, Fayle SA. Further dental treatment needs of children receiving exodontia under general anaesthesia at a teaching hospital in the UK. Int J Paediatr Dent. 2006;16(4):263-9. 10.1111/j.1365-263X.2006.00747.x
3. Foster T, Perinpanayagam H, Pfaffenbach A, et al. Recurrence of early childhood caries after comprehensive treatment with general anesthesia and follow-up. J Dent Child (Chic). 2006;73(1):25-30. PMID 16734310
4. Azadani EN, Yoshioka M, Peng J, et al. Restorative and patient factors associated with repeat general anesthesia for dental treatments in young children: A case-control study. J Am Dent Assoc. 2024;155(9):765-773. 10.1016/j.adaj.2024.06.014
5. Kirby J, Walshaw EG, Yesudian G, et al. Repeat paediatric dental general anaesthesia at Sheffield Children's NHS Foundation Trust: a service evaluation. Br Dent J. 2020;228(4):255-258. 10.1038/s41415-020-1256-9
6. Mallineni SK, Yiu CK. Dental treatment under general anesthesia for special-needs patients: analysis of the literature. J Investig Clin Dent. 2016;7(4):325-331. 10.1111/jicd.12174
7. Ashley PF, Williams CE, Moles DR, et al. Sedation versus general anaesthesia for provision of dental treatment to patients younger than 18 years. Cochrane Database Syst Rev. 2015;2015(9):CD006334. 10.1002/14651858.CD006334.pub4
8. Vacharaksa K, Thongrakard T, Vacharaksa A. Needs of non-pharmacological management versus sedation or general anesthesia for dental treatment in older adults with special needs: A systematic review. Health Policy. 2025;154:105262. 10.1016/j.healthpol.2025.105262
9. Moore DL, Ding L, Yang G, et al. Impact of Instituting General Anesthesia on Oral Sedation Care in a Tertiary Care Pediatric Dental Clinic. Anesth Prog. 2019;66(4):183-191. 10.2344/anpr-66-02-02
10. Messieha Z, Ananda RC, Hoffman I, et al. Five year outcomes study of dental rehabilitation conducted under general anesthesia for special needs patients. Anesth Prog. 2007;54(4):170-4. 10.2344/0003-3006(2007)54[170:FYOSOD]2.0.CO;2
11. American Society of Anesthesiologists Task Force on Moderate Procedural Sedation and Analgesia. Practice Guidelines for Moderate Procedural Sedation and Analgesia 2018: A Report by the American Society of Anesthesiologists Task Force on Moderate Procedural Sedation and Analgesia, the American Association of Oral and Maxillofacial Surgeons, American College of Radiology, American Dental Association, American Society of Dentist Anesthesiologists, and Society of Interventional Radiology. Anesthesiology. 2018;128(3):437-479. 10.1097/ALN.0000000000002043
12. Coté CJ, Wilson S; American Academy of Pediatrics; American Academy of Pediatric Dentistry. Guidelines for Monitoring and Management of Pediatric Patients Before, During, and After Sedation for Diagnostic and Therapeutic Procedures. Pediatrics. 2019;143(6):e20191000. 10.1542/peds.2019-1000
13. Bailey PL, Pace NL, Ashburn MA, et al. Frequent hypoxemia and apnea after sedation with midazolam and fentanyl. Anesthesiology. 1990;73(5):826-30. 10.1097/00000542-199011000-00005
14. Milgrom P, Beirne OR, Fiset L, et al. The safety and efficacy of outpatient midazolam intravenous sedation for oral surgery with and without fentanyl. Anesth Prog. 1993;40(3):57-62. PMID 7645789
15. Wiemer SJ, Mediratta JK, Triana RR, et al. What Is the Incidence of Anesthesia-Related Adverse Events in Oral and Maxillofacial Surgery Offices? A Review of 61,237 Sedation Cases From a Large Private Practice Consortium. J Oral Maxillofac Surg. 2024;82(8):895-901. 10.1016/j.joms.2024.04.014
16. Saunders R, Struys MMRF, Pollock RF, et al. Patient safety during procedural sedation using capnography monitoring: a systematic review and meta-analysis. BMJ Open. 2017;7(6):e013402. 10.1136/bmjopen-2016-013402
17. Parker W, Estrich CG, Abt E, et al. Benefits and harms of capnography during procedures involving moderate sedation: A rapid review and meta-analysis. J Am Dent Assoc. 2018;149(1):38-50.e2. 10.1016/j.adaj.2017.08.030
18. Brady P, Iohom G, O'Halloran KD, et al. Microstream capnography during conscious sedation with midazolam for oral surgery: a randomised controlled trial. BDJ Open. 2017;3:17019. 10.1038/bdjopen.2017.19
19. Takaya K, Higuchi H, Ishii-Maruhama M, et al. Capnography Prevents Hypoxia during Sedation for Dental Treatment: A Randomized Controlled Trial. JDR Clin Trans Res. 2017;2(2):158-167. 10.1177/2380084416674670
20. Fu ES, Downs JB, Schweiger JW, et al. Supplemental oxygen impairs detection of hypoventilation by pulse oximetry. Chest. 2004;126(5):1552-8. 10.1378/chest.126.5.1552
21. Hu YH, Tsai A, Ou-Yang LW, et al. Postoperative dental morbidity in children following dental treatment under general anesthesia. BMC Oral Health. 2018;18(1):84. 10.1186/s12903-018-0545-z
22. McCann ME, de Graaff JC, Dorris L, et al. Neurodevelopmental outcome at 5 years of age after general anaesthesia or awake-regional anaesthesia in infancy (GAS): an international, multicentre, randomised, controlled equivalence trial. Lancet. 2019;393(10172):664-677. 10.1016/S0140-6736(18)32485-1
23. Ünal T, Kizilay FN, Ünal SE, et al. Dental treatment outcomes under general anesthesia in pediatric patients with and without autism spectrum disorder: A retrospective analysis. Medicine (Baltimore). 2026;105(14):e46066. 10.1097/MD.0000000000046066
24. Bağ İ, Çalışkan S, Erenel AO, et al. Does the Modality of Dental Treatment Affect the Treatment Prognosis and the Necessity of Re-Treatments?. Children (Basel). 2023;10(10). 10.3390/children10101705
25. Keykha E, Hajisadeghi S, Heidari MT. The effect of dental rehabilitation under general anesthesia on dental anxiety in children: a systematic review and meta-analysis. BMC Oral Health. 2025;25(1):1953. 10.1186/s12903-025-07334-y
26. Coté CJ, Notterman DA, Karl HW, et al. Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics. 2000;105(4 Pt 1):805-14. 10.1542/peds.105.4.805