Does crowding influence extraction decisions?

One of the most controversial areas of orthodontic treatment planning involves the decision to extract teeth. This has divided our speciality for many years.  When we look at factors that influence our decision to extract, the primary influence may be clinical experience. This leads to variation between operators and perhaps the development of extraction/non-extraction philosophies along with their gurus and followers.

Currently, we appear to be going through a non-extraction phase of prescribing.  This is evident by the trend to use expansion with many different devices and the influence of “airway-friendly orthodontists” who are suggesting that extractions cause breathing disorders.  It is even suggested that orthodontic extractions are similar to amputation of body parts.

This new study attempted to identify new information that could explain orthodontists’ extraction decisions.  The study team examined whether there was an association between the amount of dental crowding and the extraction decision of a large sample of orthodontists.

What did they ask?

They did this study to

“Assess the relationship between estimated crowding and the recommendation for extraction by orthodontists in the US”.

What did they do?

They did a cross-sectional observational study. This had the following structure.

• They collected the records of four patients with Class I malocclusion with varying degrees of crowding.
• The degree of crowding was measured by a trained examiner. In addition, they traced the cephalometric radiographs.
• The team then sent an online survey to all AAO partners in the research program and the Orthodontic Pearls Facebook group.
• The survey had two main parts. Part 1 collected data on the clinician, for example, their sex, clinical background, years of experience, etc.
• The second part of the survey contained images of the patients’ records. The orthodontists were asked to estimate the amount of crowding and whether they suggested extracting teeth as part of the orthodontic treatment.
• The final stage involved running the relevant exploratory statistics and methods to assess crowding estimation and the extraction recommendation.

What did they find?

The team distributed the survey to 1904 members of the AAO partners in the research programme and 10,404 members of Orthodontic Pearls. 373 clinicians responded, and they received complete responses from 373 clinicians.

23% of the practitioners had practiced for over 5 years and 38% for over 15 years.

When they looked at the records of the 4 patients.

• Patient A had the most minor crowding and 95% of practitioners suggested non-extraction.
• Patients B and D had the most crowding and most practitioners suggested extractions (77-84%).
• Case C had crowding greater than A but less than B and D. As a result, 72% of the practitioners wanted to treat non-extraction.

Most practitioners recommended extraction once crowding reached 9-10mm in the maxilla or mandible. When they chose non-extraction, they wanted to create space by expansion, incisor proclination and interproximal reduction.

Participants with over 15 years of experience were 2.7 times more likely to suggest extraction in patient C.

Their conclusions were

“Crowding estimation varied widely between clinicians. Most clinicians recommended extraction once maxillary or mandibular crowding approximated 9-10 mm”.

What did I think?

This study followed the methods used in several other studies examining the extraction decision. In this respect, we need to consider that this was a still-life study and did not really represent real-life close interaction with our patients. Nevertheless, it does provide information that gives an insight into our treatment decisions. Furthermore, this is the only methodology we can use to evaluate our decisions.

In many ways, I feel that the findings strongly represent clinical practice. I have looked closely at the cases they included in the paper and I think that I would not extract in patients A and C. But I would extract in patients B and D. I am reassured that this seemed to be the majority decision from the panel. It is a shame that this paper is not open access so all readers can view the cases and make these decisions.

Like all studies, there are limitations to the investigation. The most important of these is that they could not calculate a response rate because they did not know how many practitioners viewed the request to participate on social media. This means that there must be some biases present in the sample.

Finally, I feel that this does help add to our knowledge as it illustrates what may be accepted practice and shows that we still do not all agree on this decision.