International Journal of Radiation Oncology * Biology * Physics
Volume 69, Issue 3, Supplement , Pages S195-S196, 1 November 2007

Impact of Advanced Technologies on the Rate of Deviations in Radiation Treatment Delivery

  • L.B. Marks

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • K.L. Light

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • D.L. Georgas

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • J.L. Hubbs

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • E.L. Jones

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • M.C. Wright

      Affiliations

    • Department of Anesthesiology, Duke University Medical Center, Durham, NC
  • ,
  • C.G. Willett

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC
  • ,
  • F. Yin

      Affiliations

    • Department of Radiation Oncology, Duke University Medical Center, Durham, NC

1103

Article Outline

 

Back to Article Outline

Purpose/Objective(s) 

External radiation therapy (RT) is a human endeavor, and thus, may not be delivered as intended. Newer RT delivery technologies, (e.g. intensity-modulated RT [IMRT] using multileaf collimators [MLC] and image guidance) are being adopted, in part, to improve accuracy of RT. This study examines the impact of new technologies on the treatment deviation rate.

Back to Article Outline

Materials/Methods 

RT treatment deviation rates were prospectively monitored during a time of technology upgrade. In Jan 2003, we had 3 linacs (ages 11–22 yrs), all with “low” technology (e.g. without MLC). In 2003–4, we upgraded to 5 new linacs, all controlled with record-and-verify; 4 were “high” technology (defined as those with MLC). The deviation rates among patients treated on “high” vs “low” tech linacs (i.e. with vs without MLC) were compared over time with a 2-tailed Fisher's exact test.

Hypotheses:

1) Safeguards inherent to “high tech” yield fewer deviations than “low tech” methods. 2) Deviation rates on “high tech” will decline with time, reflecting increased experience with new technologies.

Back to Article Outline

Results 

In 2003, the deviation rates on the “high” and “low” tech units were similar (p = 0.45), contrary to hypothesis #1. However, by 2005–6, the deviation rate was lower on the “high” vs. “low tech” units (p < 0.001), consistent with hypothesis #1. The lower deviation rate seen with “high” vs “low” tech in 2005–6 was due to

a)a decline in the “high tech” deviation rate over time (p = 0.15), suggesting a “learning curve” for the new techniques, consistent with hypothesis #2, and

b)an unexpected increase in deviations on “low tech” units during the technology upgrade (p = 0.05).

By 2006, the deviation rate on “high tech” was lower than our 2003, pre-upgrade, “low tech” deviation rate (p = 0.01). As treatment complexity is greater with “high tech” units (e.g. IMRT), the reduced deviation rate suggests safeguards inherent with “high tech” approaches are effective. None of the deviations resulted from technical malfunctions, but rather typically involved human factors: e.g. data entry, documentation, and patient set-up (Table 1).

Back to Article Outline

Conclusions 

Advanced RT delivery systems appear to reduce the rate of treatment deviations. Deviation rates on “high tech” linacs decline over time, suggesting a “learning curve” after the introduction of new technologies. The increase in deviations with “low tech” approaches was unexpected and may reflect a growing over-reliance on the tools inherent to “high tech” machines. As new technologies are introduced, continued diligence is needed to assure that staff remains proficient with “low tech” approaches. We have instituted additional procedures to reduce deviation rates for patients treated with “low” and “high” technologies.

Summary of Results: Deviation Rates, % (# of deviations/total # of treatments)
All Machines“Low” Technology“High” Technology
20030.12% (29/23,764)0.11% (23/19,962)0.16% (6/3,802)p = 0.45
20040.09% (23/24,937) 0.09% (23/24,937)
20050.13% (36/28,523)0.22% (12/5,419)0.10% (24/23,104)p = 0.0009 (“low” vs. “high” for combined 2005–2006)
20060.08% (25/31,019)0.20% (9/4,431)0.06% (16/24,857)
p = 0.053 2003 vs. 2005–2006p = 0.15 2003 vs. 2005–2006

Note: No deviations were considered clinically-meaningful, nor resulted in patient injury.

Low tech: defined somewhat arbitrarily as those machines without MLC.

High tech: defined somewhat arbitrarily as those machines with MLC.

 Author Disclosure: L.B. Marks, None; K.L. Light, None; D.L. Georgas, None; J.L. Hubbs, None; E.L. Jones, None; M.C. Wright, None; C.G. Willett, None; F. Yin, None.

PII: S0360-3016(07)01535-0

doi:10.1016/j.ijrobp.2007.07.354

International Journal of Radiation Oncology * Biology * Physics
Volume 69, Issue 3, Supplement , Pages S195-S196, 1 November 2007