Skip to main content

Artificial intelligence enhanced ultrasound (AI-US) in a severe obese parturient: a case report

Abstract

Background

Neuraxial anesthesia in obese parturients can be challenging due to anatomical and physiological modifications secondary to pregnancy; this led to growing popularity of spine ultrasound in this population for easing landmark identification and procedure execution.

Integration of Artificial Intelligence with ultrasound (AI-US) for image enhancement and analysis has increased clinicians' ability to localize vertebral structures in patients with challenging anatomical conformation.

Case presentation

We present the case of a parturient with extremely severe obesity, with a Body Mass Index (BMI) = 64.5 kg/m2, in which the AI-Enabled Image Recognition allowed a successful placing of an epidural catheter.

Conclusions

Benefits gained from AI-US implementation are multiple: immediate recognition of anatomical structures leads to increased first-attempt success rate, making easier the process of spinal anesthesia execution compared to traditional palpation methods, reducing needle placement time for spinal anesthesia and predicting best needle direction and target structure depth in peridural anesthesia.

Background

According to the WHO, excess body weight represents one of the most severe public health challenges of the twenty-first century in Europe. This issue affects more women than men and is reflected in a severe increase of obesity in pregnant women, with all the related risks [1].

Neuraxial analgesia is currently the most effective option for pain management during labor. However, in obese parturient, central neuraxial blocks can be challenging due to anatomical and physiological modifications secondary to pregnancy and the underlying disease [2].

In this population, considering the higher frequency of comorbidities and the higher risk of obstetric complications, epidural catheter placement can be a lifeline in an emergent or unplanned conversion to cesarean Sect [3].

To this end, neuraxial ultrasonography has become increasingly popular for epidural space identification. Recently, the integration of Artificial Intelligence for ultrasound image (AI-US) enhancement and analysis has further increased clinicians' ability to locate spine structures in patients with challenging anatomical conformations.

In the present case, a portable handheld AI enhanced ultrasound device played a key role in successfully placing an epidural catheter in a parturient with extreme obesity (BMI = 64.5 kg/m2), proving to be superior to palpation and conventional spine ultrasound imaging [4].

Case presentation

After obtaining written informed consent from the patient for publication of this case report and accompanying images, we present the case of a 37-year-old woman (gravida 2 para 1, gestational age 38 weeks + 5 days) who requested epidural analgesia for labor. The measured patient's height was 153 cm and her weight, on the day before delivery, was 151 kg, with a calculated BMI of 64.5 kg/m2. Past medical history was relevant for pharmacologically treated gestational hypothyroidism and diabetes during the previous pregnancy.

Manual palpation of the spine was carried out with the patient in sitting position; certain localization of interspinous spaces was not possible (Fig. 1, image c).

Fig. 1
figure 1

a Image obtained with s-US; b image obtained with AI-US; c surface anatomy of the parturient's back: no anatamical landmark can be identified with palpation

Lumbar spine ultrasound imaging was then acquired using a standard ultrasound (s-US) machine (SonoSite M-turbo®) with a convex probe (Fig. 1, image a). Insonation of the spine was performed by placing the probe midline at sacrum level, in the transverse orientation, and then shifting it cephalad to recognize intervertebral spaces and posterior and anterior complexes. Even with s-US aid, locating interspinous spaces was not feasible, owing to the quantity of subcutaneous tissue that made identifying target structures arduous.

As a last resort, we decided to employ AI-US, a dedicated handheld device combining real-time ultrasound (Fig. 1, image b) with machine learning to assist identification of anatomical structures of the spine (Accuro®, Rivanna Medical, Charlottesville, VA, USA). With it, we could identify desired intervertebral space, correct needle insertion point—marked on the patient's skin—and estimated skin-to-epidural space distance—estimated 8 cm, with a slight inclination to the left. It is important to note the inconsistency between the insertion point localized with Accuro® and that presumed after landmark palpation.

The epidural catheter was then placed by a senior attending physician, requiring a single attempt; epidural space was encountered at a 10 cm depth (with a slight discrepancy compared with AI-US esteem—most likely due to tissue compression during images acquisition). Satisfactory labor analgesia was then administered through the epidural catheter. No procedural complications are to be reported.

Conclusions

We could not find descriptions of epidural catheter placement AI-US assisted in parturient with such high BMI value in the available literature. In the case we present, morphological alterations secondary to pregnancy and obesity created difficulties that could not be overcome using traditional landmark palpation nor standard ultrasound techniques. Nevertheless, the implementation of AI-US has determined the first-step success of the procedure [5].

Preprocedural ultrasound of the patient's lumbar region helps with obtaining important information about spine anatomy: midline identification, optimal vertebral level for catheter placement, the inclination of vertebral bodies and processes and the distance from the skin to the epidural space [6].

Pre-puncture ultrasound is well-known to reduce the number of attempts and significantly increase parturients’ satisfaction in regard to the procedure [7].

This technique is even more helpful when applied in those cases with anticipated difficulty, including anatomical alteration of the lumbar spine and a body mass index (BMI) > 33 kg/m2 [8].

However, neuraxial s-US in pregnancy, especially in obese patients, can be tricky as the visibility of the ligamentum flavum, dura mater, and epidural space decreases significantly during pregnancy. In addition, the distance from the skin to the epidural space seems to increase proportionally to BMI [9].

Becoming familiar with the sonoanatomy of the spinal column requires a high level of technical expertise, so that adoption of neuraxial ultrasound has not been widespread.

In recent years, AI and machine learning-based ultrasound image analysis are gaining momentum as research subjects [4, 5, 10]. These technologies may offer a new advantage in improving outcomes and represent a training aid for operators that are not experienced in neuraxial insonation techniques [4].

Several applications of AI-US have been proposed: automatized identification of organ structures and lesions, assessment of disease status and specific categorization [11]. Two natural fields of implementation of neuraxial AI-US are obstetric and orthopaedic anesthesia. Automated landmark identification programs have been shown effective in identifying needle insertion points in obese pregnant women requiring spinal anaesthesia for cesarean Sect.  [5] as well as in epidural catheter positioning in parturients requesting labor epidural analgesia and in combined spinal–epidural anaesthesia for cesarean delivery, showing positive impact on increasing first-attempt success and shortening procedure's duration [4, 10].

When performing spinal anaesthesia in obese patients undergoing orthopaedic procedures, anesthesiologists needed to redirect the needle fewer times when AI-US was implemented. Of note, interspinous spaces identified as per digital palpation has been shown to be less precise when compared to AI-US; this inconsistency was also particularly evident in our case [12].

In conclusion, benefits brought to the field by AI-US are multiple, all reflected in significantly increased patient satisfaction. In both spinal and epidural anesthesia, AI-US increases efficacy of interspinous space location, reduces needle placement time and predicts needle direction for reaching of target structures as well their distances from skin [13, 14].

Neuroaxial s-US is an advanced skill that relies on the operator for providing accurate results.

When compared to s-US, AI-US provides the clinicians more detailed information that can be pivotal in more complex clinical scenarios. In Table 1 are summarized strengths and core features of both techniques.

Table 1 Strengths of different neuraxial ultrasound methods

There is still much room for improvement and we are far from considering AI-US the standard for neuraxial anaesthesia. When ultrasound became available for practical use at the bedside, it led to a change in our clinical practice, for instance, in the way we look at vascular access and at peripheral nerve blocks. This historical turning point came not smoothly. Clinical trials and accumulation of experience and expertise were needed to make practitioners accept the novelties. We do not know if AI-US will become the new paradigm in neuraxial ultrasound. However, we do think it is a powerful tool we must start considering in our algorithms as well as for further investigations, systematic studies on this subject are warranted.

Availability of data and materials

The data sets used and/or analysed during the current study are available from the corresponding author on request.

Abbreviations

s-US:

Standard ultrasound

AI-US:

Artificial intelligence enhanced ultrasound

BMI:

Body max index

References

  1. Catalano PM, Shankar K (2017) Obesity and pregnancy: mechanisms of short term and long term adverse consequences for mother and child. BMJ. https://doi.org/10.1136/bmj.j1

    Article  PubMed  PubMed Central  Google Scholar 

  2. Patel SD, Habib AS (2021) Anaesthesia for the parturient with obesity. BJA Education. https://doi.org/10.1016/j.bjae.2020.12.007

    Article  PubMed  PubMed Central  Google Scholar 

  3. Kim ST (2021) Anesthetic management of obese and morbidly obese parturients. Anesth Pain Med. https://doi.org/10.17085/apm.21090

    Article  PubMed  PubMed Central  Google Scholar 

  4. Ni X et al (2021) Accuro ultrasound-based system with computer-aided image interpretation compared to traditional palpation technique for neuraxial anesthesia placement in obese parturients undergoing cesarean delivery: a randomized controlled trial. J Anesth 35:475–482

    Article  Google Scholar 

  5. In Chan JJ et al (2021) Machine learning approach to needle insertion site identification for spinal anesthesia in obese patients. BMC Anesthesiol. https://doi.org/10.1186/s12871-021-01466-8

    Article  PubMed  PubMed Central  Google Scholar 

  6. National Institute for Health and Care Excellence. (2008). Ultrasound-guided catheterisation of the epidural space. NICE Guidel.

  7. Grau T, Leipold RW, Conradi R, Martin E (2001) Ultrasound control for presumed difficult epidural puncture. Acta Anaesthesiol Scand. https://doi.org/10.1034/j.1399-6576.2001.045006766.x

    Article  PubMed  Google Scholar 

  8. Yoo S, Kim Y, Park S-K, Ji S-H, Kim J-T (2020) Ultrasonography for lumbar neuraxial block. Anesth Pain Med. https://doi.org/10.17085/apm.20065

    Article  Google Scholar 

  9. Clinkscales CP, Greenfield MLVH, Vanarase M, Polley LS (2007) An observational study of the relationship between lumbar epidural space depth and body mass index in Michigan parturients. Int J Obste Anesth. https://doi.org/10.1016/j.ijoa.2007.03.015

    Article  Google Scholar 

  10. Seligman KM, Weiniger CF, Carvalho B (2018) The accuracy of a handheld ultrasound device for neuraxial depth and landmark assessment. Anesth Analg 126:1995–1998

    Article  Google Scholar 

  11. Shen YT, Chen L, Yue WW, Xu HX (2021) Artificial intelligence in ultrasound. Eur J Radio. https://doi.org/10.1016/j.ejrad.2021.109717

    Article  Google Scholar 

  12. Ghisi D et al (2020) A randomized comparison between Accuro and palpation-guided spinal anesthesia for obese patients undergoing orthopedic surgery. Reg Anesth Pain Med 45:63–66

    Article  Google Scholar 

  13. Tiouririne M, Dixon AJ, Mauldin FW, Scalzo D, Krishnaraj A (2017) Imaging performance of a handheld ultrasound system with real-time computer-aided detection of lumbar spine anatomy: a feasibility study. Invest Radiol 52:447–455

    Article  Google Scholar 

  14. Singla P et al (2019) Feasibility of spinal anesthesia placement using automated interpretation of lumbar ultrasound images: a prospective randomized controlled trial. J Clin Anesth Res. https://doi.org/10.4172/2155-6148.1000878

    Article  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

The authors have no sources of funding to declare for this manuscript. Rivanna Medical provided the device used for the study but had no role in study design or manuscript write-up.

Author information

Authors and Affiliations

Authors

Contributions

GB, AC and CC: Investigation and Resources; GB, AC and MT: Writing—Original Draft; CC, VB and EB: Review & Editing and Supervision; All authors read and approved the final manuscript.

Corresponding author

Correspondence to Alberto Calabrese.

Ethics declarations

Ethics approval and consent to participate

This type of study is exempted from ethical approval in our institution.

Consent for publication

We obtained verbal and written informed consent from the patient for this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Competing interests

The authors declare no conflicts of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Compagnone, C., Borrini, G., Calabrese, A. et al. Artificial intelligence enhanced ultrasound (AI-US) in a severe obese parturient: a case report. Ultrasound J 14, 34 (2022). https://doi.org/10.1186/s13089-022-00283-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13089-022-00283-5

Keywords