Why cache MongoDB connection in Next.js? And does it work?

I'm creating a Next.js application and I noticed that many developers cache the MongoDB connection. For example

let cachedClient = null;
let cachedDb = null;

export async function connectToDatabase() {
    if (cachedClient && cachedDb) {
        return {
            client: cachedClient,
            db: cachedDb,
        };
    }

    const opts = {
        useNewUrlParser: true,
        useUnifiedTopology: true,
    };

    let client = new MongoClient(MONGODB_URI, opts);
    await client.connect();
    let db = client.db(MONGODB_DB);

    cachedClient = client;
    cachedDb = db;

    return {
        client: cachedClient,
        db: cachedDb,
    };
}

or

let cached = global.mongoose

if (!cached) {
  cached = global.mongoose = { conn: null, promise: null }
}

async function dbConnect () {
  if (cached.conn) {
    return cached.conn
  }

  if (!cached.promise) {
    const opts = {
      useNewUrlParser: true,
      useUnifiedTopology: true,
      bufferCommands: false,
      bufferMaxEntries: 0,
      useFindAndModify: true,
      useCreateIndex: true
    }

    cached.promise = mongoose.connect(MONGODB_URI, opts).then(mongoose => {
      return mongoose
    })
  }
  cached.conn = await cached.promise
  return cached.conn
}

I've never seen that in Express apllications so I have 2 questions:

1. Why is caching database connection such a common thing in Next.js while I've never seen that in Express.js. What's he reason for that? How does it work? And is it worth it?
2. As you can see in the examples above some developer use useual let-variables while some other developer use global variables. What's the difference and which is the better solution?