Abstract
Background
Long-term follow-up is essential for assessing the efficacy of surgical methods in pediatric patients. However, cohort dropouts tend to increase over time. These losses to follow-up make it difficult to obtain reliable and convincing results. The aim of this article is to present an easily understandable graphical method for visualizing the postoperative course using a cohort of patients with congenital blepharoptosis.
Methods
Patients with congenital blepharoptosis who underwent frontalis suspension were retrospectively reviewed. The postoperative courses of all the patients were illustrated using two modified swimmer plots to depict reoperation events. One plot used postoperative years as the x-axis, while the other used patients' ages. Each horizontal bar represented an individual patient, with reoperations indicated by changes in color.
Results
A total of 130 patients was included. The plot with postoperative years as the x-axis showed that most reoperations occurred >9 years after the initial surgery. The age-based plot showed that most secondary revision surgeries were performed between 10 and 20 years of age. The age distribution of reoperations followed an approximately sigmoidal curve, with a median age of 15.6 years. This distribution resembled the follow-up endpoints of patients who did not undergo reoperation.
Conclusions
A modified swimmer plot that includes all patients is a useful tool for visualizing longitudinal outcomes and identifying issues that may be overlooked with conventional analytical methods.
Keywords: Swimmer plot, Frontalis suspension, Congenital blepharoptosis, Long-term evaluation
To assess the efficacy of surgical methods in pediatric patients, long-term follow-up is essential. However, cohort dropouts increase as the study duration extends. These losses to follow-up hinder the ability to obtain reliable and convincing results.1 The aim of this article is to present a graphical method using a cohort of patients with congenital blepharoptosis.
Methods
Patients with congenital blepharoptosis who underwent frontalis suspension with fascia lata transplantation at Osaka City General Hospital between April 2000 and December 2010 were retrospectively reviewed. Patients younger than 10 years at the time of their first operation were included. The operative procedure followed the technique reported by Sakahara.2 The collected patient characteristics included gender, laterality of the affected eye, age at first surgery, whether revision operations were performed, age at revision surgery, and the last recorded follow-up date. Descriptive statistics were calculated. Postoperative courses for all patients were illustrated using two modified swimmer plots to depict reoperation events. One plot was based on postoperative years, while the other was based on patients' ages. Each horizontal bar represented an individual patient, and reoperations were indicated by changes in color.
Results
A total of 130 patients (89 males and 41 females) was included. The median age at the first operation was 4.8 years. Fifty-one patients were affected on the right side, 56 on the left, and 23 on both sides. Secondary revision surgery was performed in 22 patients. All of these procedures involved repositioning of the transplanted fascia, except for one case that required fascia replantation. Among these 22 patients, 8 underwent a third operation, including one case of replantation. The graph with postoperative years on the x-axis showed that most reoperations were performed >9 years after the initial surgery (Figure. 1). The median time to reoperation was 10.6 years, with an interquartile range of 9.7 years for the first quartile and 11.4 years for the third quartile. The other graph, in which the x-axis represents patient age, showed that most secondary surgeries were performed between 10 and 20 years of age. The age distribution of reoperations followed an approximately sigmoidal curve, with a median age of 15.6 years. This distribution was similar to the follow-up endpoints of patients who did not undergo reoperation (Figure. 2).
Figure. 1.
Postoperative years and reoperations. Each horizontal bar represents an individual patient, and reoperations are indicated by changes in color. Nearly all reoperations occurred >9 years after the initial surgery. OP, operation.
Figure. 2.
Patients’ ages and reoperations. The age distribution of reoperations follows an approximately sigmoidal curve, which closely resembles the follow-up endpoints of patients who did not undergo reoperation. OP, operation.
Discussion
Long-term follow-up is important for evaluating surgical procedures in children. However, conventional methods make it difficult to illustrate the time course of postoperative outcomes. In the literature, the minimum follow-up period is typically around six months.3 Conversely, extended observation periods inevitably lead to increased dropout rates, which make the results more vulnerable to bias. A previous long-term follow-up report from the authors’ institution also included fewer than half of the original cases.2
Presenting data in figures or graphs allows complex information to be rapidly interpreted, facilitating an overall understanding. In the field of chemotherapy, swimmer plots are used to depict the fluctuating relationship between tumor response and patient survival.4 This graphical approach appears suitable for displaying the postoperative course, as it incorporates the trajectory of all patients and enables visual assessment of patient dropouts.
To simplify the graph, the investigation focused on reoperations, which generally imply suboptimal outcomes. Whether revision surgery was performed could be judged objectively from the medical records, requiring minimal discretion from the investigators.
When evaluating the outcomes of a surgical method in pediatric patients, both the patient's age and the postoperative duration are equally important. Therefore, one plot was modified to reflect patient age, while the other illustrated the postoperative time course.
The plot with postoperative years on the x-axis showed that few secondary operations occurred within the first 5 years, whereas the proportion of revisions increased markedly after 10 years, highlighting the importance of long-term follow-up.
The graph sorted by patient age showed that most reoperations were performed between 10 and 20 years of age. The age distribution of reoperations was similar to the follow-up endpoints of patients who did not undergo reoperation. In light of this, follow-up until at least ten years of age is necessary for comprehensive evaluation.
Investigating the factors associated with revision surgery is beyond the scope of this article. However, social factors, in addition to medical causes, may contribute to secondary operations. Further research is needed to reach definitive conclusions.
In conclusion, graphically representing the longitudinal postoperative course of all cases is a valuable tool for sharing insights into outcomes and for identifying issues that may be overlooked with conventional analysis.
Declaration of competing interest
The author declares no conflict of interest, and that no funding was received for this research or for the preparation of the manuscript.
Ethical Approval
Received from the local IRB (2306042).
References
- 1.Dettori J.R. Loss to follow-up. Evid Based Spine Care J. 2011;2:7–10. doi: 10.1055/s-0030-1267080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sakahara D., Imai K., Masuoka T., et al. Long-term functional and aesthetic evaluation of using a single rhomboid-shaped fascial strip for severe congenital unilateral blepharoptosis. J Plast Reconstruct Aesthet Surg. 2018;71:1618–1624. doi: 10.1016/j.bjps.2018.07.003. [DOI] [PubMed] [Google Scholar]
- 3.Ueno K., Shirakawa Y., Uemura K., Kumegawa S., Wada Y., Asamura S. Surgical results of retroauricular fascia graft in frontalis suspension surgery: a case series study. J Plast Reconstruct Aesthet Surg. 2024;88:266–272. doi: 10.1016/j.bjps.2023.10.133. [DOI] [PubMed] [Google Scholar]
- 4.Chia P.L., Gedye C., Boutros P.C., et al. Current and evolving methods to visualize biological data in cancer research. J Natl Cancer Inst. 2016;108:djw031. doi: 10.1093/jnci/djw031. [DOI] [PMC free article] [PubMed] [Google Scholar]