Joshua B. Kayser and Paul N. Lanken

A Primer on Data Collection

In the modern ICU, large amounts of patient data are generated on a daily basis and need timely review and evaluation. These include not only the data at hand but also what may be referred to as meta-data—that is, trends or other changes in today’s data compared to data from yesterday and prior days and patterns of changes in the current data compared to prior patterns (Box 13.1). As a result, ICU clinicians are required to assimilate and codify an extraordinary number of details for each of their ICU patients in order to make decisions and organize a plan of care.

Three processes are involved in memory: encoding, storage, and retrieval. Encoding refers to how something is processed for memory in the brain. Once it has been encoded, it can then be stored in the form of a short-term or long-term memory. Retrieval is the process of getting information from a memory. Most ICU data are stored in a subdivision of the short-term memory, termed the working memory, for quick processing. However, humans can only process a limited number of details (about seven) for short intervals in the working memory.

However, a typical ICU clinician is required to be familiar with many more than seven details daily for each ICU patient. This presents an intrinsic challenge to successful processing of ICU data on rounds and day-to-day management of critically ill patients, especially when the ICU clinician must also follow trends or evolving patterns of changes in data. Additionally, the data are often subject to irregular sampling, measurement error, and interpretation error, as well as the inherent bias of the individual clinician in terms of what to believe and base decisions on. Accuracy and consistency can thus be difficult to achieve.

Monitoring of Overnight Events and Patient Assessment

In this age of medical care, information is passed from clinician to clinician with greater frequency than in the past (e.g., so-called handoffs or handovers). It is therefore crucial that a detailed account of overnight events be given at the beginning of each day. A recommended approach is for the daytime ICU clinicians to discuss major events from the previous night with the physicians and nurses who cover the nighttime shift, followed by a review of notes and documentation. ICU nurses spend the largest percentage of time at the bedside of their patients, and are thus an invaluable resource. Input from clinicians based on telemedicine units that remotely monitor certain ICU patients (Chapter 111) can also be an important source of information about the nighttime events.

The daily assessment proceeds with systematic review of vital signs and fluid balance (i.e., intake and output—I’s and O’s) over the previous 24 hours. In the ICU, vital signs include temperature, blood pressure, pulse, respirations, oxygen saturation (by pulse oximetry), and pain and other important signs and symptoms (e.g., level of sedation and presence of delirium).

In evaluating the patient’s temperature curve, awareness of both hyperthermic and hypothermic episodes can provide useful information about infectious and inflammatory conditions in addition to common complications of ICU stays such as atelectasis and drug reactions.

Blood pressure is of vital importance to maintain normal body homeostasis. Adequate blood pressure is a key component in enabling the body to successfully deliver oxygen to the tissue and cellular level. Assessment of blood pressure should include both a review of systolic and diastolic pressures as well as the mean arterial pressure, which can serve as an indirect assessment of organ perfusion pressure.

Review of the patient’s pulse should include both a quantitative assessment of heart rate as well as a qualitative assessment of rhythm. Additionally, it is important to review the alarm history on the telemetry monitor to diagnose any arrhythmias that may have occurred over the interval of interest.

Assessment of respiratory status and arterial oxygen saturation should include both breathing rate and pattern. Awareness of abnormal rates can be useful to diagnose increased work of breathing, which may subsequently lead to respiratory failure, or very slow rates, which may result from oversedation with respiratory depressant medications. Unusual patterns of respirations, such as Cheyne-Stokes or Kussmaul breathing or episodes of obstructive apneas, can also be valuable in diagnosing a patient’s pathology.

Patients with respiratory failure who require mechanical ventilation should have their ventilator mode and relevant settings reviewed as well as serial arterial blood gases (ABGs) to assess acid base status, ventilation, and oxygenation. The latter includes assessing how closely the pulse oximetry readings of O₂ saturation are to the calculated or measured O₂ saturation by ABGs or co-oximetry, respectively. In association with respirations and O₂ saturations and ABGs, ventilator settings and functions should be reviewed at the bedside, including tidal volume (total and mL/kg predicted body weight [PBW], Appendix E), airway pressures (peak and plateau), minute ventilation, and evidence of auto positive end expiratory pressure (auto-PEEP) (see Chapters 2, 3, and 47).

Because of its importance for patient- and family-centered care, some argue that pain measurement (e.g., Figure 5.1 in Chapter 5 and Figure 87.1 in Chapter 87) should be regarded and treated as the “fifth vital sign” in ICU patients. The same could be argued regarding the patient’s level of consciousness (e.g., level of sedation or agitation and the presence and severity of delirium). Regarding the latter, it is recommended that both the goal of sedative therapy and actual level of sedation achieved be communicated by a standard method, such as by using the Richmond Agitation-Sedation Scale (RASS) (Chapter 5). Likewise, a standard method of evaluating for the presence of delirium (e.g., the Confusion Assessment Method for Intensive Care Unit [CAM-ICU]) is preferred over less systematic methods.

Lastly, volume status (intravascular and total body fluid volumes) should be reviewed. Assessment of total intake and output, including a breakdown in type of intake (e.g., parenteral versus enteral) and output (urine, stool, drains, and tubes) can help clarify the significance of imbalances between “ins” and “outs.” Urine output, in particular, is a simple way to assess for adequate organ perfusion (with usual threshold of adequate urine output being 0.5 mL/kg PBW/h). If patients have indwelling central catheters, notation of accurate measurements of central venous pressure (CVP) or pulmonary arterial wedge pressure (PAWP) can aid in the assessment of volume status.

Once the assessment of vital signs is complete, providers should examine the patient and talk with the patient and family members, if present (Chapter 104). At a minimum, one should perform a focused physical exam, including breath and heart sounds, mental status, and other neurologic signs, and inquire about the presence and intensity of pain, dyspnea, and other symptoms. When doing one’s physical exam, one should carefully evaluate the skin to measure skin turgor; identify new pressure ulcers (Chapter 42), ecchymoses, or rashes (Chapter 43); assess temperature and capillary refill time in all extremities; and look for signs of infection at the insertion site of medical devices (Chapters 11 and 14). A mental status exam should be tailored to the individual patient, with either a qualitative assessment (alert, delirious, somnolent, obtunded, etc.) or a quantitative assessment, such as the Glasgow Coma Scale (GCS) (Chapter 99), a sedation scale (e.g., RASS) (Chapter 5), or a delirium assessment (e.g., CAM-ICU) (Chapter 37).

A review of the ICU flow sheet (paper or electronic) can be helpful for assessing changes in clinical status of patients over the past 24 hours or longer as documented by the patient’s nurses and respiratory therapists. Likewise, a review of the medical record (paper or electronic) for results of diagnostic tests in the past 24 hours as well as notes and recommendations by consultants, house staff, or other members of the ICU clinical team is recommended to round out the daily picture.

Medication Reconciliation and Nutritional Management

Examination of intravenous (IV) fluids and drugs that are being administered as continous IV infusions (“drips”) and other medications is an essential part of the daily patient assessment. Common IV drips in an ICU setting include IV fluids, vasopressors, sedatives, analgesics, and antimicrobials. Pertinent information about IV drips includes the type and rate, changes over the past 24 hours or longer, as well as whether an IV drug is being administered continuously or by bolus. When assessing vasopressors, it is important to note any trends of dosage changes that may reflect a change in the patient’s hemodynamic status. For patients on sedation, awareness of whether patients are on continuous infusions or bolus or if they receive a daily sedation interruption (i.e., a spontaneous awakening trial SAT, see Chapter 5) and its results are key components of the assessment of mental status, as sedation can adversely impact a patient’s degree of alertness and cognition (see Chapter 36).

Antimicrobials should be reviewed in a systematic manner to avoid overutilization, which can result in antimicrobial resistance. When possible, a daily therapeutic plan should exist for each antimicrobial, including knowledge of the rationale for its use, the current number of days of therapy, and the planned number of days of treatment.

All medication orders should be reviewed daily, and any medications that are judged to be no longer necessary should be discontinued. Lastly, knowledge of other medications should include awareness of the dose and frequency as well as an assessment of administration (e.g., being administered, and if not, why being held) and awareness of potential drug side effects and interactions.

As with medications, the patient’s current nutritional support should be reviewed in terms of type (Chapter 15), route of administration (Chapter 16), and whether or not it’s at the nutritional goal as recommended by nutritional consultants. How the patient is tolerating the current level of nutritional support (e.g., presence of gastric residuals, abdominal distension and discomfort, and presence and type of stool) is an important element to keep on top of as well as whether the nutritional therapy is having its desired effect in terms of the patient’s body weight and malnutrition indices (e.g., albumin and prealbumin) starting to trend in the appropriate directions of improved nutrition status (Chapter 15).

Laboratory Data

Critically ill patients routinely have an enormous amount of laboratory data. First, as a general imperative for all hospitalized patients, one should not order a laboratory study (“lab”) if it is not needed. Just because patients are in the ICU doesn’t necessarily mean they must have all or even most of their labs drawn daily.

Commonly, labs fall into one of three categories: metabolic, cellular, or coagulation. The most common metabolic studies include the basic metabolic panel (BMP) or “panel 7,” which includes the serum sodium (Na⁺), potassium (K⁺), chloride (Cl⁻), and bicarbonate (CO₂⁻) as well as serum blood urea nitrogen (BUN), creatinine (Cr), and glucose (Glu). The complete or comprehensive metabolic panel (CMP) includes BMP elements plus serum calcium (Ca⁺), albumin (Alb⁻), phosphate (P), and liver function tests (LFTs). Assessment of electrolytes is useful to review water and acid/base balance, renal function, and glycemic control. LFTs and albumin aid in evaluating hepatic function and the patient’s nutritional status.

The cellular test most commonly ordered is the complete blood count (CBC), which allows for assessment of the white blood cell (WBC), hemoglobin and hematocrit (H&H or Hgb & Hct), and platelet (Plt) count. A review of the WBC should include not only total WBC count (leukocytosis or leukopenia) but notation of type of circulating WBC (polymorphoneutrophils [PMNs or “polys”], immature PMNs or bands, lymphocytes, etc.). A decrease in H&H can help to explain subjective dyspnea, pallor, new blood loss, or derangements in oxygen delivery. Finally, a review of the platelet count and its trend from ICU admission can help to explain new bleeding or ecchymoses or signs of a drug-induced thrombocytopenia (Chapter 45).

Coagulation studies most commonly involve measurement of prothrombin time (PT/INR) and partial thromboplastin time (PTT), standard measurements of extrinsic and intrinsic clotting function.

Other labs may prove helpful in certain clinical circumstances. These include lactic acid levels and central venous assessments of oxygen saturation (Scv_O2), both of which relate to how well tissues are oxygenated and perfused, especially in shock states (see Chapters 8, 9, and 10), as well as microbiologic data. When reviewing microbiologic data, it is essential to be aware of the source and date of the lab as well as results, including culture and sensitivities. As an example, if a patient with a fever underwent blood cultures, the provider should know the site of the culture (e.g., peripheral right arm), the organism (e.g., Staphylococcus aureus), and the sensitivities (e.g., pan-resistant except to vancomycin). As described in Chapter 14, one should avoid drawing blood cultures through an indwelling catheter because such blood cultures have high false-positive rates (unless the cultures are taken when the catheter is freshly placed under aseptic conditions).

Other Studies

As with laboratory data, ICU patients commonly undergo a considerable number of other diagnostic studies such as chest radiographs (CXRs) and computed tomography (CT) scans. As a rule of thumb, ICU providers should be aware of all study results of radiological studies performed or interpreted during the interval since the prior daily rounds and be prepared to discuss them on the current rounds.

Falling Urine Output and Rising Creatinine

With rare exception, a falling urine output (UOP) typically heralds the onset of acute renal failure (ARF) or, under more current terminology, acute kidney injury (AKI) (Chapter 81). AKI is a common occurrence in the ICU, with up to one third of patients experiencing some degree of AKI during an ICU admission. The drop in UOP is typically a result of decreases in the glomerular filtration rate (GFR) and is associated with the accumulation of urea, creatinine, and body fluids. The differential diagnosis for a rising creatinine and BUN in ICU patients is shown in Table 13.E1. These, in turn, result in various clinical abnormalities, e.g., changes in mental status, electrolyte abnormalities, derangements in acid-based balance, and volume overload. The presence of AKI can have a profound impact on ICU care and outcomes, resulting in prolonged ICU and hospital stays as well as a higher risk of death (~50% of ICU patients who develop AKI in the ICU die). Additionally, it can result in substantial increases in cost. However, renal failure is typically a reversible process, provided an underlying etiology can be determined.

The kidneys have three major functions: filtration of the blood to eliminate metabolic waste, solute and acid-base balance, and volume management. AKI can therefore result in a loss of the ability to regulate any of those responsibilities. AKI typically occurs over a period of hours to days and can occur de novo or in addition to underlying chronic renal dysfunction. Renal failure can be defined as nonoliguric (> 400 mL UOP/day), oliguric (< 400 mL UOP/day), and anuric (< 100 mL UOP/day). All three result in variable reductions in the ability to maintain the necessary renal functions.

There are three classifications of AKI: prerenal, intrarenal (intrinsic), and postrenal. Briefly, prerenal AKI results from renal hypoperfusion, intrinsic AKI results from parenchymal disease, and postrenal AKI results from urinary tract obstruction (Box 13.E1).

Prerenal AKI is the most common of the three and occurs because of a deficit in effective intravascular circulating volume. Therefore anything that reduces renal perfusion can precipitate prerenal AKI. Examples include hypotension, volume depletion (e.g., dehydration, hemorrhage, gastrointestinal losses, burns, renal losses, increased insensible loses), redistribution of volume (e.g., capillary leak syndromes, vasodilation, disorders of oncotic/hydrostatic pressure), cardiac dysfunction (e.g., congestive heart failure [CHF]), and medication effects on the renal vasculature (e.g., nonsteroidal anti-inflammatory drugs [NSAIDs], angiotensin-converting enzyme [ACE] inhibitors).

Intrinsic AKI results from disorders that affect the renal parenchyma, including the vasculature, glomerulus, interstitium, and tubules. The most common form of intrarenal AKI is acute tubular necrosis (ATN). ATN has multiple etiologies, including ischemia and exposure to nephrotoxic medications. Common nephrotoxic drugs encountered in the ICU include NSAIDs, aminoglycosides, angiotensin converting enzyme inhibitors (ACE-I) and angiotensin receptor blockers